diff --git "a/trainer_state.json" "b/trainer_state.json" new file mode 100644--- /dev/null +++ "b/trainer_state.json" @@ -0,0 +1,35033 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.48828125, + "eval_steps": 500, + "global_step": 5000, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 9.765625e-05, + "grad_norm": 13.125093460083008, + "learning_rate": 1.6666666666666669e-06, + "loss": 6.0473, + "step": 1 + }, + { + "epoch": 0.0001953125, + "grad_norm": 13.109691619873047, + "learning_rate": 3.3333333333333337e-06, + "loss": 6.0151, + "step": 2 + }, + { + "epoch": 0.00029296875, + "grad_norm": 12.975897789001465, + "learning_rate": 5e-06, + "loss": 6.0374, + "step": 3 + }, + { + "epoch": 0.000390625, + "grad_norm": 12.861470222473145, + "learning_rate": 6.6666666666666675e-06, + "loss": 5.9534, + "step": 4 + }, + { + "epoch": 0.00048828125, + "grad_norm": 11.551366806030273, + "learning_rate": 8.333333333333334e-06, + "loss": 5.7805, + "step": 5 + }, + { + "epoch": 0.0005859375, + "grad_norm": 7.968364238739014, + "learning_rate": 1e-05, + "loss": 5.5664, + "step": 6 + }, + { + "epoch": 0.00068359375, + "grad_norm": 6.854646682739258, + "learning_rate": 1.1666666666666668e-05, + "loss": 5.4982, + "step": 7 + }, + { + "epoch": 0.00078125, + "grad_norm": 3.759798049926758, + "learning_rate": 1.3333333333333335e-05, + "loss": 5.1824, + "step": 8 + }, + { + "epoch": 0.00087890625, + "grad_norm": 3.48822283744812, + "learning_rate": 1.5e-05, + "loss": 5.0667, + "step": 9 + }, + { + "epoch": 0.0009765625, + "grad_norm": 3.1011886596679688, + "learning_rate": 1.6666666666666667e-05, + "loss": 5.0028, + "step": 10 + }, + { + "epoch": 0.00107421875, + "grad_norm": 2.5868849754333496, + "learning_rate": 1.8333333333333333e-05, + "loss": 4.7787, + "step": 11 + }, + { + "epoch": 0.001171875, + "grad_norm": 2.378497838973999, + "learning_rate": 2e-05, + "loss": 4.6855, + "step": 12 + }, + { + "epoch": 0.00126953125, + "grad_norm": 2.2101738452911377, + "learning_rate": 2.1666666666666667e-05, + "loss": 4.6042, + "step": 13 + }, + { + "epoch": 0.0013671875, + "grad_norm": 1.9938961267471313, + "learning_rate": 2.3333333333333336e-05, + "loss": 4.3843, + "step": 14 + }, + { + "epoch": 0.00146484375, + "grad_norm": 1.9345463514328003, + "learning_rate": 2.5e-05, + "loss": 4.2548, + "step": 15 + }, + { + "epoch": 0.0015625, + "grad_norm": 1.7730776071548462, + "learning_rate": 2.666666666666667e-05, + "loss": 4.1363, + "step": 16 + }, + { + "epoch": 0.00166015625, + "grad_norm": 1.58334481716156, + "learning_rate": 2.8333333333333332e-05, + "loss": 3.9704, + "step": 17 + }, + { + "epoch": 0.0017578125, + "grad_norm": 1.4827001094818115, + "learning_rate": 3e-05, + "loss": 3.81, + "step": 18 + }, + { + "epoch": 0.00185546875, + "grad_norm": 1.4140430688858032, + "learning_rate": 3.166666666666667e-05, + "loss": 3.7305, + "step": 19 + }, + { + "epoch": 0.001953125, + "grad_norm": 1.3934496641159058, + "learning_rate": 3.3333333333333335e-05, + "loss": 3.5375, + "step": 20 + }, + { + "epoch": 0.00205078125, + "grad_norm": 1.3384771347045898, + "learning_rate": 3.5000000000000004e-05, + "loss": 3.5217, + "step": 21 + }, + { + "epoch": 0.0021484375, + "grad_norm": 1.2228978872299194, + "learning_rate": 3.6666666666666666e-05, + "loss": 3.3742, + "step": 22 + }, + { + "epoch": 0.00224609375, + "grad_norm": 1.1775028705596924, + "learning_rate": 3.8333333333333334e-05, + "loss": 3.2626, + "step": 23 + }, + { + "epoch": 0.00234375, + "grad_norm": 1.0886842012405396, + "learning_rate": 4e-05, + "loss": 3.1864, + "step": 24 + }, + { + "epoch": 0.00244140625, + "grad_norm": 1.0981535911560059, + "learning_rate": 4.1666666666666665e-05, + "loss": 3.0962, + "step": 25 + }, + { + "epoch": 0.0025390625, + "grad_norm": 1.009294867515564, + "learning_rate": 4.3333333333333334e-05, + "loss": 3.0507, + "step": 26 + }, + { + "epoch": 0.00263671875, + "grad_norm": 0.9390103816986084, + "learning_rate": 4.4999999999999996e-05, + "loss": 2.9579, + "step": 27 + }, + { + "epoch": 0.002734375, + "grad_norm": 0.8647847175598145, + "learning_rate": 4.666666666666667e-05, + "loss": 2.8166, + "step": 28 + }, + { + "epoch": 0.00283203125, + "grad_norm": 0.8606237769126892, + "learning_rate": 4.8333333333333334e-05, + "loss": 2.7919, + "step": 29 + }, + { + "epoch": 0.0029296875, + "grad_norm": 0.8069576025009155, + "learning_rate": 5e-05, + "loss": 2.7426, + "step": 30 + }, + { + "epoch": 0.00302734375, + "grad_norm": 0.7007808685302734, + "learning_rate": 5.1666666666666664e-05, + "loss": 2.6926, + "step": 31 + }, + { + "epoch": 0.003125, + "grad_norm": 0.6825646162033081, + "learning_rate": 5.333333333333334e-05, + "loss": 2.6783, + "step": 32 + }, + { + "epoch": 0.00322265625, + "grad_norm": 0.7054751515388489, + "learning_rate": 5.5e-05, + "loss": 2.6239, + "step": 33 + }, + { + "epoch": 0.0033203125, + "grad_norm": 0.7101700305938721, + "learning_rate": 5.6666666666666664e-05, + "loss": 2.5993, + "step": 34 + }, + { + "epoch": 0.00341796875, + "grad_norm": 0.56828773021698, + "learning_rate": 5.833333333333333e-05, + "loss": 2.5258, + "step": 35 + }, + { + "epoch": 0.003515625, + "grad_norm": 0.5513983368873596, + "learning_rate": 6e-05, + "loss": 2.5018, + "step": 36 + }, + { + "epoch": 0.00361328125, + "grad_norm": 0.573391318321228, + "learning_rate": 6.166666666666667e-05, + "loss": 2.4881, + "step": 37 + }, + { + "epoch": 0.0037109375, + "grad_norm": 0.4520920515060425, + "learning_rate": 6.333333333333335e-05, + "loss": 2.4489, + "step": 38 + }, + { + "epoch": 0.00380859375, + "grad_norm": 0.43038809299468994, + "learning_rate": 6.500000000000001e-05, + "loss": 2.4387, + "step": 39 + }, + { + "epoch": 0.00390625, + "grad_norm": 0.4071808457374573, + "learning_rate": 6.666666666666667e-05, + "loss": 2.4261, + "step": 40 + }, + { + "epoch": 0.00400390625, + "grad_norm": 0.3631410300731659, + "learning_rate": 6.833333333333333e-05, + "loss": 2.3656, + "step": 41 + }, + { + "epoch": 0.0041015625, + "grad_norm": 0.33457818627357483, + "learning_rate": 7.000000000000001e-05, + "loss": 2.4011, + "step": 42 + }, + { + "epoch": 0.00419921875, + "grad_norm": 0.3582305610179901, + "learning_rate": 7.166666666666667e-05, + "loss": 2.3662, + "step": 43 + }, + { + "epoch": 0.004296875, + "grad_norm": 0.4090467691421509, + "learning_rate": 7.333333333333333e-05, + "loss": 2.3282, + "step": 44 + }, + { + "epoch": 0.00439453125, + "grad_norm": 0.34438556432724, + "learning_rate": 7.5e-05, + "loss": 2.2991, + "step": 45 + }, + { + "epoch": 0.0044921875, + "grad_norm": 0.29381102323532104, + "learning_rate": 7.666666666666667e-05, + "loss": 2.2506, + "step": 46 + }, + { + "epoch": 0.00458984375, + "grad_norm": 0.4019562304019928, + "learning_rate": 7.833333333333334e-05, + "loss": 2.2494, + "step": 47 + }, + { + "epoch": 0.0046875, + "grad_norm": 0.3183911144733429, + "learning_rate": 8e-05, + "loss": 2.2853, + "step": 48 + }, + { + "epoch": 0.00478515625, + "grad_norm": 0.33041393756866455, + "learning_rate": 8.166666666666667e-05, + "loss": 2.2251, + "step": 49 + }, + { + "epoch": 0.0048828125, + "grad_norm": 0.28936123847961426, + "learning_rate": 8.333333333333333e-05, + "loss": 2.2308, + "step": 50 + }, + { + "epoch": 0.00498046875, + "grad_norm": 0.36125779151916504, + "learning_rate": 8.5e-05, + "loss": 2.2473, + "step": 51 + }, + { + "epoch": 0.005078125, + "grad_norm": 0.2832348942756653, + "learning_rate": 8.666666666666667e-05, + "loss": 2.201, + "step": 52 + }, + { + "epoch": 0.00517578125, + "grad_norm": 0.3250133693218231, + "learning_rate": 8.833333333333333e-05, + "loss": 2.1596, + "step": 53 + }, + { + "epoch": 0.0052734375, + "grad_norm": 0.31502828001976013, + "learning_rate": 8.999999999999999e-05, + "loss": 2.2183, + "step": 54 + }, + { + "epoch": 0.00537109375, + "grad_norm": 0.33068645000457764, + "learning_rate": 9.166666666666667e-05, + "loss": 2.2254, + "step": 55 + }, + { + "epoch": 0.00546875, + "grad_norm": 0.30684661865234375, + "learning_rate": 9.333333333333334e-05, + "loss": 2.2572, + "step": 56 + }, + { + "epoch": 0.00556640625, + "grad_norm": 0.31540846824645996, + "learning_rate": 9.5e-05, + "loss": 2.1594, + "step": 57 + }, + { + "epoch": 0.0056640625, + "grad_norm": 0.23259752988815308, + "learning_rate": 9.666666666666667e-05, + "loss": 2.172, + "step": 58 + }, + { + "epoch": 0.00576171875, + "grad_norm": 0.26666638255119324, + "learning_rate": 9.833333333333333e-05, + "loss": 2.1961, + "step": 59 + }, + { + "epoch": 0.005859375, + "grad_norm": 0.26295527815818787, + "learning_rate": 0.0001, + "loss": 2.1811, + "step": 60 + }, + { + "epoch": 0.00595703125, + "grad_norm": 0.2302207201719284, + "learning_rate": 0.00010166666666666667, + "loss": 2.1431, + "step": 61 + }, + { + "epoch": 0.0060546875, + "grad_norm": 0.468537837266922, + "learning_rate": 0.00010333333333333333, + "loss": 2.133, + "step": 62 + }, + { + "epoch": 0.00615234375, + "grad_norm": 0.27946797013282776, + "learning_rate": 0.000105, + "loss": 2.1917, + "step": 63 + }, + { + "epoch": 0.00625, + "grad_norm": 0.26789286732673645, + "learning_rate": 0.00010666666666666668, + "loss": 2.1417, + "step": 64 + }, + { + "epoch": 0.00634765625, + "grad_norm": 0.22977805137634277, + "learning_rate": 0.00010833333333333334, + "loss": 2.1371, + "step": 65 + }, + { + "epoch": 0.0064453125, + "grad_norm": 0.2590682804584503, + "learning_rate": 0.00011, + "loss": 2.113, + "step": 66 + }, + { + "epoch": 0.00654296875, + "grad_norm": 0.2660805583000183, + "learning_rate": 0.00011166666666666667, + "loss": 2.1346, + "step": 67 + }, + { + "epoch": 0.006640625, + "grad_norm": 0.24538874626159668, + "learning_rate": 0.00011333333333333333, + "loss": 2.1078, + "step": 68 + }, + { + "epoch": 0.00673828125, + "grad_norm": 0.2986001968383789, + "learning_rate": 0.000115, + "loss": 2.1048, + "step": 69 + }, + { + "epoch": 0.0068359375, + "grad_norm": 0.30193910002708435, + "learning_rate": 0.00011666666666666667, + "loss": 2.135, + "step": 70 + }, + { + "epoch": 0.00693359375, + "grad_norm": 0.25153177976608276, + "learning_rate": 0.00011833333333333334, + "loss": 2.1101, + "step": 71 + }, + { + "epoch": 0.00703125, + "grad_norm": 0.7685809135437012, + "learning_rate": 0.00012, + "loss": 2.0833, + "step": 72 + }, + { + "epoch": 0.00712890625, + "grad_norm": 0.24468126893043518, + "learning_rate": 0.00012166666666666668, + "loss": 2.1409, + "step": 73 + }, + { + "epoch": 0.0072265625, + "grad_norm": 0.3070752024650574, + "learning_rate": 0.00012333333333333334, + "loss": 2.1402, + "step": 74 + }, + { + "epoch": 0.00732421875, + "grad_norm": 0.352273553609848, + "learning_rate": 0.000125, + "loss": 2.171, + "step": 75 + }, + { + "epoch": 0.007421875, + "grad_norm": 0.34806111454963684, + "learning_rate": 0.0001266666666666667, + "loss": 2.1155, + "step": 76 + }, + { + "epoch": 0.00751953125, + "grad_norm": 0.317414253950119, + "learning_rate": 0.00012833333333333333, + "loss": 2.0812, + "step": 77 + }, + { + "epoch": 0.0076171875, + "grad_norm": 0.3320539891719818, + "learning_rate": 0.00013000000000000002, + "loss": 2.0987, + "step": 78 + }, + { + "epoch": 0.00771484375, + "grad_norm": 0.3315901756286621, + "learning_rate": 0.00013166666666666665, + "loss": 2.1025, + "step": 79 + }, + { + "epoch": 0.0078125, + "grad_norm": 0.29021286964416504, + "learning_rate": 0.00013333333333333334, + "loss": 2.1474, + "step": 80 + }, + { + "epoch": 0.00791015625, + "grad_norm": 0.2593044936656952, + "learning_rate": 0.000135, + "loss": 2.0667, + "step": 81 + }, + { + "epoch": 0.0080078125, + "grad_norm": 0.3192054033279419, + "learning_rate": 0.00013666666666666666, + "loss": 2.0764, + "step": 82 + }, + { + "epoch": 0.00810546875, + "grad_norm": 0.26053664088249207, + "learning_rate": 0.00013833333333333333, + "loss": 2.0698, + "step": 83 + }, + { + "epoch": 0.008203125, + "grad_norm": 0.24790963530540466, + "learning_rate": 0.00014000000000000001, + "loss": 2.124, + "step": 84 + }, + { + "epoch": 0.00830078125, + "grad_norm": 0.2546316087245941, + "learning_rate": 0.00014166666666666668, + "loss": 2.1025, + "step": 85 + }, + { + "epoch": 0.0083984375, + "grad_norm": 0.22275258600711823, + "learning_rate": 0.00014333333333333334, + "loss": 2.0778, + "step": 86 + }, + { + "epoch": 0.00849609375, + "grad_norm": 0.22855599224567413, + "learning_rate": 0.000145, + "loss": 2.1113, + "step": 87 + }, + { + "epoch": 0.00859375, + "grad_norm": 0.2456902116537094, + "learning_rate": 0.00014666666666666666, + "loss": 2.1399, + "step": 88 + }, + { + "epoch": 0.00869140625, + "grad_norm": 0.22144420444965363, + "learning_rate": 0.00014833333333333335, + "loss": 2.0211, + "step": 89 + }, + { + "epoch": 0.0087890625, + "grad_norm": 0.254894495010376, + "learning_rate": 0.00015, + "loss": 2.1382, + "step": 90 + }, + { + "epoch": 0.00888671875, + "grad_norm": 0.2729082703590393, + "learning_rate": 0.00015166666666666668, + "loss": 2.1271, + "step": 91 + }, + { + "epoch": 0.008984375, + "grad_norm": 0.2564642131328583, + "learning_rate": 0.00015333333333333334, + "loss": 2.0423, + "step": 92 + }, + { + "epoch": 0.00908203125, + "grad_norm": 0.24927419424057007, + "learning_rate": 0.000155, + "loss": 2.0603, + "step": 93 + }, + { + "epoch": 0.0091796875, + "grad_norm": 0.2591923475265503, + "learning_rate": 0.0001566666666666667, + "loss": 2.0522, + "step": 94 + }, + { + "epoch": 0.00927734375, + "grad_norm": 0.3350547254085541, + "learning_rate": 0.00015833333333333332, + "loss": 2.1095, + "step": 95 + }, + { + "epoch": 0.009375, + "grad_norm": 0.30227839946746826, + "learning_rate": 0.00016, + "loss": 2.121, + "step": 96 + }, + { + "epoch": 0.00947265625, + "grad_norm": 0.3027198314666748, + "learning_rate": 0.00016166666666666665, + "loss": 2.0288, + "step": 97 + }, + { + "epoch": 0.0095703125, + "grad_norm": 0.29791951179504395, + "learning_rate": 0.00016333333333333334, + "loss": 2.0471, + "step": 98 + }, + { + "epoch": 0.00966796875, + "grad_norm": 0.30918803811073303, + "learning_rate": 0.000165, + "loss": 2.1036, + "step": 99 + }, + { + "epoch": 0.009765625, + "grad_norm": 0.2510216534137726, + "learning_rate": 0.00016666666666666666, + "loss": 2.0217, + "step": 100 + }, + { + "epoch": 0.00986328125, + "grad_norm": 0.288231760263443, + "learning_rate": 0.00016833333333333335, + "loss": 2.0178, + "step": 101 + }, + { + "epoch": 0.0099609375, + "grad_norm": 0.3326691687107086, + "learning_rate": 0.00017, + "loss": 2.0947, + "step": 102 + }, + { + "epoch": 0.01005859375, + "grad_norm": 0.3552946150302887, + "learning_rate": 0.00017166666666666667, + "loss": 1.9953, + "step": 103 + }, + { + "epoch": 0.01015625, + "grad_norm": 0.34882935881614685, + "learning_rate": 0.00017333333333333334, + "loss": 2.1223, + "step": 104 + }, + { + "epoch": 0.01025390625, + "grad_norm": 0.35487979650497437, + "learning_rate": 0.000175, + "loss": 2.0599, + "step": 105 + }, + { + "epoch": 0.0103515625, + "grad_norm": 0.33561578392982483, + "learning_rate": 0.00017666666666666666, + "loss": 2.0398, + "step": 106 + }, + { + "epoch": 0.01044921875, + "grad_norm": 0.27415409684181213, + "learning_rate": 0.00017833333333333335, + "loss": 2.0982, + "step": 107 + }, + { + "epoch": 0.010546875, + "grad_norm": 0.28592920303344727, + "learning_rate": 0.00017999999999999998, + "loss": 2.0571, + "step": 108 + }, + { + "epoch": 0.01064453125, + "grad_norm": 0.3228552043437958, + "learning_rate": 0.00018166666666666667, + "loss": 1.9923, + "step": 109 + }, + { + "epoch": 0.0107421875, + "grad_norm": 0.29005661606788635, + "learning_rate": 0.00018333333333333334, + "loss": 2.0394, + "step": 110 + }, + { + "epoch": 0.01083984375, + "grad_norm": 0.36677825450897217, + "learning_rate": 0.000185, + "loss": 2.0153, + "step": 111 + }, + { + "epoch": 0.0109375, + "grad_norm": 0.2562806308269501, + "learning_rate": 0.0001866666666666667, + "loss": 2.05, + "step": 112 + }, + { + "epoch": 0.01103515625, + "grad_norm": 0.2748093008995056, + "learning_rate": 0.00018833333333333332, + "loss": 2.0449, + "step": 113 + }, + { + "epoch": 0.0111328125, + "grad_norm": 0.3924459218978882, + "learning_rate": 0.00019, + "loss": 2.0409, + "step": 114 + }, + { + "epoch": 0.01123046875, + "grad_norm": 0.4368191063404083, + "learning_rate": 0.00019166666666666667, + "loss": 2.0178, + "step": 115 + }, + { + "epoch": 0.011328125, + "grad_norm": 0.4236035943031311, + "learning_rate": 0.00019333333333333333, + "loss": 2.0518, + "step": 116 + }, + { + "epoch": 0.01142578125, + "grad_norm": 0.28912851214408875, + "learning_rate": 0.00019500000000000002, + "loss": 2.0058, + "step": 117 + }, + { + "epoch": 0.0115234375, + "grad_norm": 0.38847798109054565, + "learning_rate": 0.00019666666666666666, + "loss": 2.0818, + "step": 118 + }, + { + "epoch": 0.01162109375, + "grad_norm": 0.29429811239242554, + "learning_rate": 0.00019833333333333335, + "loss": 2.004, + "step": 119 + }, + { + "epoch": 0.01171875, + "grad_norm": 0.2969271242618561, + "learning_rate": 0.0002, + "loss": 2.046, + "step": 120 + }, + { + "epoch": 0.01181640625, + "grad_norm": 0.297894150018692, + "learning_rate": 0.00020166666666666667, + "loss": 2.0204, + "step": 121 + }, + { + "epoch": 0.0119140625, + "grad_norm": 0.28794237971305847, + "learning_rate": 0.00020333333333333333, + "loss": 2.0663, + "step": 122 + }, + { + "epoch": 0.01201171875, + "grad_norm": 0.2766033411026001, + "learning_rate": 0.000205, + "loss": 2.0399, + "step": 123 + }, + { + "epoch": 0.012109375, + "grad_norm": 0.3516612648963928, + "learning_rate": 0.00020666666666666666, + "loss": 2.0389, + "step": 124 + }, + { + "epoch": 0.01220703125, + "grad_norm": 0.28204381465911865, + "learning_rate": 0.00020833333333333335, + "loss": 2.0084, + "step": 125 + }, + { + "epoch": 0.0123046875, + "grad_norm": 0.3306240737438202, + "learning_rate": 0.00021, + "loss": 2.013, + "step": 126 + }, + { + "epoch": 0.01240234375, + "grad_norm": 0.3221111595630646, + "learning_rate": 0.00021166666666666667, + "loss": 2.0312, + "step": 127 + }, + { + "epoch": 0.0125, + "grad_norm": 0.2742249071598053, + "learning_rate": 0.00021333333333333336, + "loss": 2.0375, + "step": 128 + }, + { + "epoch": 0.01259765625, + "grad_norm": 0.2785228490829468, + "learning_rate": 0.000215, + "loss": 2.0449, + "step": 129 + }, + { + "epoch": 0.0126953125, + "grad_norm": 0.2666397988796234, + "learning_rate": 0.00021666666666666668, + "loss": 2.0355, + "step": 130 + }, + { + "epoch": 0.01279296875, + "grad_norm": 0.22522135078907013, + "learning_rate": 0.00021833333333333332, + "loss": 1.9693, + "step": 131 + }, + { + "epoch": 0.012890625, + "grad_norm": 0.2724483013153076, + "learning_rate": 0.00022, + "loss": 2.0671, + "step": 132 + }, + { + "epoch": 0.01298828125, + "grad_norm": 0.2655040919780731, + "learning_rate": 0.00022166666666666667, + "loss": 1.9805, + "step": 133 + }, + { + "epoch": 0.0130859375, + "grad_norm": 0.3194504678249359, + "learning_rate": 0.00022333333333333333, + "loss": 2.067, + "step": 134 + }, + { + "epoch": 0.01318359375, + "grad_norm": 0.25511813163757324, + "learning_rate": 0.00022500000000000002, + "loss": 2.0141, + "step": 135 + }, + { + "epoch": 0.01328125, + "grad_norm": 0.3995087444782257, + "learning_rate": 0.00022666666666666666, + "loss": 2.0278, + "step": 136 + }, + { + "epoch": 0.01337890625, + "grad_norm": 0.2623380422592163, + "learning_rate": 0.00022833333333333334, + "loss": 1.9751, + "step": 137 + }, + { + "epoch": 0.0134765625, + "grad_norm": 0.23814889788627625, + "learning_rate": 0.00023, + "loss": 1.969, + "step": 138 + }, + { + "epoch": 0.01357421875, + "grad_norm": 0.23566491901874542, + "learning_rate": 0.00023166666666666667, + "loss": 2.0135, + "step": 139 + }, + { + "epoch": 0.013671875, + "grad_norm": 0.2437373697757721, + "learning_rate": 0.00023333333333333333, + "loss": 2.0044, + "step": 140 + }, + { + "epoch": 0.01376953125, + "grad_norm": 0.2861543595790863, + "learning_rate": 0.000235, + "loss": 2.0378, + "step": 141 + }, + { + "epoch": 0.0138671875, + "grad_norm": 0.318050354719162, + "learning_rate": 0.00023666666666666668, + "loss": 2.0857, + "step": 142 + }, + { + "epoch": 0.01396484375, + "grad_norm": 0.39669227600097656, + "learning_rate": 0.00023833333333333334, + "loss": 2.0535, + "step": 143 + }, + { + "epoch": 0.0140625, + "grad_norm": 0.4359401762485504, + "learning_rate": 0.00024, + "loss": 2.0432, + "step": 144 + }, + { + "epoch": 0.01416015625, + "grad_norm": 0.4532039165496826, + "learning_rate": 0.00024166666666666667, + "loss": 1.9909, + "step": 145 + }, + { + "epoch": 0.0142578125, + "grad_norm": 0.4570695161819458, + "learning_rate": 0.00024333333333333336, + "loss": 2.0123, + "step": 146 + }, + { + "epoch": 0.01435546875, + "grad_norm": 0.36623403429985046, + "learning_rate": 0.000245, + "loss": 2.0179, + "step": 147 + }, + { + "epoch": 0.014453125, + "grad_norm": 0.3069714307785034, + "learning_rate": 0.0002466666666666667, + "loss": 2.0014, + "step": 148 + }, + { + "epoch": 0.01455078125, + "grad_norm": 0.3980304002761841, + "learning_rate": 0.0002483333333333333, + "loss": 2.0489, + "step": 149 + }, + { + "epoch": 0.0146484375, + "grad_norm": 0.31907564401626587, + "learning_rate": 0.00025, + "loss": 2.013, + "step": 150 + }, + { + "epoch": 0.01474609375, + "grad_norm": 0.2952549159526825, + "learning_rate": 0.00025166666666666664, + "loss": 2.0709, + "step": 151 + }, + { + "epoch": 0.01484375, + "grad_norm": 0.29451197385787964, + "learning_rate": 0.0002533333333333334, + "loss": 1.9613, + "step": 152 + }, + { + "epoch": 0.01494140625, + "grad_norm": 0.2893507778644562, + "learning_rate": 0.000255, + "loss": 2.0048, + "step": 153 + }, + { + "epoch": 0.0150390625, + "grad_norm": 0.24850639700889587, + "learning_rate": 0.00025666666666666665, + "loss": 2.0198, + "step": 154 + }, + { + "epoch": 0.01513671875, + "grad_norm": 0.24297639727592468, + "learning_rate": 0.00025833333333333334, + "loss": 2.0561, + "step": 155 + }, + { + "epoch": 0.015234375, + "grad_norm": 0.2777438461780548, + "learning_rate": 0.00026000000000000003, + "loss": 2.0206, + "step": 156 + }, + { + "epoch": 0.01533203125, + "grad_norm": 0.28714093565940857, + "learning_rate": 0.00026166666666666667, + "loss": 1.9891, + "step": 157 + }, + { + "epoch": 0.0154296875, + "grad_norm": 0.25796255469322205, + "learning_rate": 0.0002633333333333333, + "loss": 2.0369, + "step": 158 + }, + { + "epoch": 0.01552734375, + "grad_norm": 0.2391008883714676, + "learning_rate": 0.00026500000000000004, + "loss": 2.0015, + "step": 159 + }, + { + "epoch": 0.015625, + "grad_norm": 0.3203892409801483, + "learning_rate": 0.0002666666666666667, + "loss": 2.0213, + "step": 160 + }, + { + "epoch": 0.01572265625, + "grad_norm": 0.3396870791912079, + "learning_rate": 0.0002683333333333333, + "loss": 2.0333, + "step": 161 + }, + { + "epoch": 0.0158203125, + "grad_norm": 0.315060555934906, + "learning_rate": 0.00027, + "loss": 2.0195, + "step": 162 + }, + { + "epoch": 0.01591796875, + "grad_norm": 0.2672436535358429, + "learning_rate": 0.0002716666666666667, + "loss": 1.9946, + "step": 163 + }, + { + "epoch": 0.016015625, + "grad_norm": 0.2996402382850647, + "learning_rate": 0.00027333333333333333, + "loss": 2.0112, + "step": 164 + }, + { + "epoch": 0.01611328125, + "grad_norm": 0.2894189953804016, + "learning_rate": 0.000275, + "loss": 2.0157, + "step": 165 + }, + { + "epoch": 0.0162109375, + "grad_norm": 0.26241254806518555, + "learning_rate": 0.00027666666666666665, + "loss": 2.0177, + "step": 166 + }, + { + "epoch": 0.01630859375, + "grad_norm": 0.22900305688381195, + "learning_rate": 0.00027833333333333334, + "loss": 1.9834, + "step": 167 + }, + { + "epoch": 0.01640625, + "grad_norm": 0.2373427301645279, + "learning_rate": 0.00028000000000000003, + "loss": 1.9792, + "step": 168 + }, + { + "epoch": 0.01650390625, + "grad_norm": 0.2663004696369171, + "learning_rate": 0.00028166666666666666, + "loss": 2.0203, + "step": 169 + }, + { + "epoch": 0.0166015625, + "grad_norm": 0.31653544306755066, + "learning_rate": 0.00028333333333333335, + "loss": 2.0216, + "step": 170 + }, + { + "epoch": 0.01669921875, + "grad_norm": 0.3077234923839569, + "learning_rate": 0.000285, + "loss": 2.0361, + "step": 171 + }, + { + "epoch": 0.016796875, + "grad_norm": 0.25555703043937683, + "learning_rate": 0.0002866666666666667, + "loss": 2.0102, + "step": 172 + }, + { + "epoch": 0.01689453125, + "grad_norm": 0.29817435145378113, + "learning_rate": 0.0002883333333333333, + "loss": 1.972, + "step": 173 + }, + { + "epoch": 0.0169921875, + "grad_norm": 0.3075692355632782, + "learning_rate": 0.00029, + "loss": 2.0195, + "step": 174 + }, + { + "epoch": 0.01708984375, + "grad_norm": 0.29917964339256287, + "learning_rate": 0.0002916666666666667, + "loss": 1.9972, + "step": 175 + }, + { + "epoch": 0.0171875, + "grad_norm": 0.32018229365348816, + "learning_rate": 0.0002933333333333333, + "loss": 1.9895, + "step": 176 + }, + { + "epoch": 0.01728515625, + "grad_norm": 0.2907097339630127, + "learning_rate": 0.000295, + "loss": 1.9777, + "step": 177 + }, + { + "epoch": 0.0173828125, + "grad_norm": 0.22390642762184143, + "learning_rate": 0.0002966666666666667, + "loss": 1.9961, + "step": 178 + }, + { + "epoch": 0.01748046875, + "grad_norm": 0.25350186228752136, + "learning_rate": 0.00029833333333333334, + "loss": 1.9875, + "step": 179 + }, + { + "epoch": 0.017578125, + "grad_norm": 0.22856706380844116, + "learning_rate": 0.0003, + "loss": 2.0169, + "step": 180 + }, + { + "epoch": 0.01767578125, + "grad_norm": 0.2288493812084198, + "learning_rate": 0.0003016666666666667, + "loss": 2.0238, + "step": 181 + }, + { + "epoch": 0.0177734375, + "grad_norm": 0.27326855063438416, + "learning_rate": 0.00030333333333333335, + "loss": 2.0134, + "step": 182 + }, + { + "epoch": 0.01787109375, + "grad_norm": 0.2447524517774582, + "learning_rate": 0.000305, + "loss": 1.9852, + "step": 183 + }, + { + "epoch": 0.01796875, + "grad_norm": 0.4363366663455963, + "learning_rate": 0.0003066666666666667, + "loss": 1.9921, + "step": 184 + }, + { + "epoch": 0.01806640625, + "grad_norm": 0.4569666385650635, + "learning_rate": 0.00030833333333333337, + "loss": 1.9997, + "step": 185 + }, + { + "epoch": 0.0181640625, + "grad_norm": 0.43348655104637146, + "learning_rate": 0.00031, + "loss": 2.0584, + "step": 186 + }, + { + "epoch": 0.01826171875, + "grad_norm": 0.3844921588897705, + "learning_rate": 0.00031166666666666663, + "loss": 2.0035, + "step": 187 + }, + { + "epoch": 0.018359375, + "grad_norm": 0.3427641987800598, + "learning_rate": 0.0003133333333333334, + "loss": 1.9536, + "step": 188 + }, + { + "epoch": 0.01845703125, + "grad_norm": 0.33557865023612976, + "learning_rate": 0.000315, + "loss": 1.972, + "step": 189 + }, + { + "epoch": 0.0185546875, + "grad_norm": 0.4006612300872803, + "learning_rate": 0.00031666666666666665, + "loss": 2.0652, + "step": 190 + }, + { + "epoch": 0.01865234375, + "grad_norm": 0.3158099055290222, + "learning_rate": 0.00031833333333333334, + "loss": 2.0516, + "step": 191 + }, + { + "epoch": 0.01875, + "grad_norm": 0.3799190819263458, + "learning_rate": 0.00032, + "loss": 2.011, + "step": 192 + }, + { + "epoch": 0.01884765625, + "grad_norm": 0.2948876619338989, + "learning_rate": 0.00032166666666666666, + "loss": 2.0109, + "step": 193 + }, + { + "epoch": 0.0189453125, + "grad_norm": 0.24561335146427155, + "learning_rate": 0.0003233333333333333, + "loss": 2.0264, + "step": 194 + }, + { + "epoch": 0.01904296875, + "grad_norm": 0.24896866083145142, + "learning_rate": 0.00032500000000000004, + "loss": 1.9726, + "step": 195 + }, + { + "epoch": 0.019140625, + "grad_norm": 0.26887547969818115, + "learning_rate": 0.0003266666666666667, + "loss": 2.0036, + "step": 196 + }, + { + "epoch": 0.01923828125, + "grad_norm": 0.3186735212802887, + "learning_rate": 0.0003283333333333333, + "loss": 2.0174, + "step": 197 + }, + { + "epoch": 0.0193359375, + "grad_norm": 0.3317165672779083, + "learning_rate": 0.00033, + "loss": 2.0047, + "step": 198 + }, + { + "epoch": 0.01943359375, + "grad_norm": 0.3068574070930481, + "learning_rate": 0.0003316666666666667, + "loss": 2.0032, + "step": 199 + }, + { + "epoch": 0.01953125, + "grad_norm": 0.29292526841163635, + "learning_rate": 0.0003333333333333333, + "loss": 2.0533, + "step": 200 + }, + { + "epoch": 0.01962890625, + "grad_norm": 0.2519834041595459, + "learning_rate": 0.000335, + "loss": 2.0113, + "step": 201 + }, + { + "epoch": 0.0197265625, + "grad_norm": 0.25766584277153015, + "learning_rate": 0.0003366666666666667, + "loss": 2.0278, + "step": 202 + }, + { + "epoch": 0.01982421875, + "grad_norm": 0.2704983055591583, + "learning_rate": 0.00033833333333333334, + "loss": 1.9725, + "step": 203 + }, + { + "epoch": 0.019921875, + "grad_norm": 0.2882053256034851, + "learning_rate": 0.00034, + "loss": 2.0706, + "step": 204 + }, + { + "epoch": 0.02001953125, + "grad_norm": 0.34524375200271606, + "learning_rate": 0.00034166666666666666, + "loss": 1.9868, + "step": 205 + }, + { + "epoch": 0.0201171875, + "grad_norm": 0.3718552887439728, + "learning_rate": 0.00034333333333333335, + "loss": 1.9596, + "step": 206 + }, + { + "epoch": 0.02021484375, + "grad_norm": 0.2747247815132141, + "learning_rate": 0.000345, + "loss": 2.0125, + "step": 207 + }, + { + "epoch": 0.0203125, + "grad_norm": 0.3062858283519745, + "learning_rate": 0.00034666666666666667, + "loss": 2.0224, + "step": 208 + }, + { + "epoch": 0.02041015625, + "grad_norm": 0.32505863904953003, + "learning_rate": 0.00034833333333333336, + "loss": 2.0376, + "step": 209 + }, + { + "epoch": 0.0205078125, + "grad_norm": 0.35048386454582214, + "learning_rate": 0.00035, + "loss": 2.0268, + "step": 210 + }, + { + "epoch": 0.02060546875, + "grad_norm": 0.31204426288604736, + "learning_rate": 0.0003516666666666667, + "loss": 2.0198, + "step": 211 + }, + { + "epoch": 0.020703125, + "grad_norm": 0.24253524839878082, + "learning_rate": 0.0003533333333333333, + "loss": 2.028, + "step": 212 + }, + { + "epoch": 0.02080078125, + "grad_norm": 0.286915123462677, + "learning_rate": 0.000355, + "loss": 1.9358, + "step": 213 + }, + { + "epoch": 0.0208984375, + "grad_norm": 0.2800680994987488, + "learning_rate": 0.0003566666666666667, + "loss": 1.9905, + "step": 214 + }, + { + "epoch": 0.02099609375, + "grad_norm": 0.2718358635902405, + "learning_rate": 0.00035833333333333333, + "loss": 1.9902, + "step": 215 + }, + { + "epoch": 0.02109375, + "grad_norm": 0.28583604097366333, + "learning_rate": 0.00035999999999999997, + "loss": 1.991, + "step": 216 + }, + { + "epoch": 0.02119140625, + "grad_norm": 0.2911478281021118, + "learning_rate": 0.0003616666666666667, + "loss": 1.973, + "step": 217 + }, + { + "epoch": 0.0212890625, + "grad_norm": 0.3601188361644745, + "learning_rate": 0.00036333333333333335, + "loss": 1.9727, + "step": 218 + }, + { + "epoch": 0.02138671875, + "grad_norm": 0.2888337969779968, + "learning_rate": 0.000365, + "loss": 1.988, + "step": 219 + }, + { + "epoch": 0.021484375, + "grad_norm": 0.25628700852394104, + "learning_rate": 0.00036666666666666667, + "loss": 1.9984, + "step": 220 + }, + { + "epoch": 0.02158203125, + "grad_norm": 0.2637641429901123, + "learning_rate": 0.00036833333333333336, + "loss": 2.0229, + "step": 221 + }, + { + "epoch": 0.0216796875, + "grad_norm": 0.23845899105072021, + "learning_rate": 0.00037, + "loss": 1.9985, + "step": 222 + }, + { + "epoch": 0.02177734375, + "grad_norm": 0.28519535064697266, + "learning_rate": 0.00037166666666666663, + "loss": 2.0061, + "step": 223 + }, + { + "epoch": 0.021875, + "grad_norm": 0.31845173239707947, + "learning_rate": 0.0003733333333333334, + "loss": 2.0081, + "step": 224 + }, + { + "epoch": 0.02197265625, + "grad_norm": 0.3725838363170624, + "learning_rate": 0.000375, + "loss": 2.0032, + "step": 225 + }, + { + "epoch": 0.0220703125, + "grad_norm": 0.49783870577812195, + "learning_rate": 0.00037666666666666664, + "loss": 2.0404, + "step": 226 + }, + { + "epoch": 0.02216796875, + "grad_norm": 0.5059479475021362, + "learning_rate": 0.0003783333333333334, + "loss": 2.0498, + "step": 227 + }, + { + "epoch": 0.022265625, + "grad_norm": 0.461291640996933, + "learning_rate": 0.00038, + "loss": 2.0078, + "step": 228 + }, + { + "epoch": 0.02236328125, + "grad_norm": 0.3970203399658203, + "learning_rate": 0.00038166666666666666, + "loss": 1.9966, + "step": 229 + }, + { + "epoch": 0.0224609375, + "grad_norm": 0.3155679404735565, + "learning_rate": 0.00038333333333333334, + "loss": 2.0263, + "step": 230 + }, + { + "epoch": 0.02255859375, + "grad_norm": 0.32979920506477356, + "learning_rate": 0.00038500000000000003, + "loss": 1.9967, + "step": 231 + }, + { + "epoch": 0.02265625, + "grad_norm": 0.27470117807388306, + "learning_rate": 0.00038666666666666667, + "loss": 2.0461, + "step": 232 + }, + { + "epoch": 0.02275390625, + "grad_norm": 0.2981088161468506, + "learning_rate": 0.0003883333333333333, + "loss": 1.9944, + "step": 233 + }, + { + "epoch": 0.0228515625, + "grad_norm": 0.3496599495410919, + "learning_rate": 0.00039000000000000005, + "loss": 1.9907, + "step": 234 + }, + { + "epoch": 0.02294921875, + "grad_norm": 0.3318106532096863, + "learning_rate": 0.0003916666666666667, + "loss": 2.0576, + "step": 235 + }, + { + "epoch": 0.023046875, + "grad_norm": 0.29498377442359924, + "learning_rate": 0.0003933333333333333, + "loss": 2.0242, + "step": 236 + }, + { + "epoch": 0.02314453125, + "grad_norm": 0.2970214784145355, + "learning_rate": 0.000395, + "loss": 2.0087, + "step": 237 + }, + { + "epoch": 0.0232421875, + "grad_norm": 0.37431418895721436, + "learning_rate": 0.0003966666666666667, + "loss": 2.0657, + "step": 238 + }, + { + "epoch": 0.02333984375, + "grad_norm": 0.30095174908638, + "learning_rate": 0.00039833333333333333, + "loss": 2.0217, + "step": 239 + }, + { + "epoch": 0.0234375, + "grad_norm": 0.24695053696632385, + "learning_rate": 0.0004, + "loss": 1.9833, + "step": 240 + }, + { + "epoch": 0.02353515625, + "grad_norm": 0.2923540771007538, + "learning_rate": 0.00040166666666666665, + "loss": 2.0272, + "step": 241 + }, + { + "epoch": 0.0236328125, + "grad_norm": 0.2788209915161133, + "learning_rate": 0.00040333333333333334, + "loss": 2.0104, + "step": 242 + }, + { + "epoch": 0.02373046875, + "grad_norm": 0.2529614567756653, + "learning_rate": 0.00040500000000000003, + "loss": 2.003, + "step": 243 + }, + { + "epoch": 0.023828125, + "grad_norm": 0.2551966905593872, + "learning_rate": 0.00040666666666666667, + "loss": 2.001, + "step": 244 + }, + { + "epoch": 0.02392578125, + "grad_norm": 0.2613292634487152, + "learning_rate": 0.00040833333333333336, + "loss": 1.9822, + "step": 245 + }, + { + "epoch": 0.0240234375, + "grad_norm": 0.3060430884361267, + "learning_rate": 0.00041, + "loss": 2.0024, + "step": 246 + }, + { + "epoch": 0.02412109375, + "grad_norm": 0.33755916357040405, + "learning_rate": 0.0004116666666666667, + "loss": 2.0023, + "step": 247 + }, + { + "epoch": 0.02421875, + "grad_norm": 0.33021774888038635, + "learning_rate": 0.0004133333333333333, + "loss": 1.9086, + "step": 248 + }, + { + "epoch": 0.02431640625, + "grad_norm": 0.26662060618400574, + "learning_rate": 0.000415, + "loss": 2.009, + "step": 249 + }, + { + "epoch": 0.0244140625, + "grad_norm": 0.27698251605033875, + "learning_rate": 0.0004166666666666667, + "loss": 2.0183, + "step": 250 + }, + { + "epoch": 0.02451171875, + "grad_norm": 0.2582184970378876, + "learning_rate": 0.00041833333333333333, + "loss": 1.9932, + "step": 251 + }, + { + "epoch": 0.024609375, + "grad_norm": 0.28684699535369873, + "learning_rate": 0.00042, + "loss": 2.0021, + "step": 252 + }, + { + "epoch": 0.02470703125, + "grad_norm": 0.33535540103912354, + "learning_rate": 0.0004216666666666667, + "loss": 2.0387, + "step": 253 + }, + { + "epoch": 0.0248046875, + "grad_norm": 0.3330588638782501, + "learning_rate": 0.00042333333333333334, + "loss": 1.9776, + "step": 254 + }, + { + "epoch": 0.02490234375, + "grad_norm": 0.27919256687164307, + "learning_rate": 0.000425, + "loss": 2.0111, + "step": 255 + }, + { + "epoch": 0.025, + "grad_norm": 0.25296416878700256, + "learning_rate": 0.0004266666666666667, + "loss": 1.9755, + "step": 256 + }, + { + "epoch": 0.02509765625, + "grad_norm": 0.31288138031959534, + "learning_rate": 0.00042833333333333335, + "loss": 2.032, + "step": 257 + }, + { + "epoch": 0.0251953125, + "grad_norm": 0.3360923230648041, + "learning_rate": 0.00043, + "loss": 2.0837, + "step": 258 + }, + { + "epoch": 0.02529296875, + "grad_norm": 0.36317816376686096, + "learning_rate": 0.0004316666666666667, + "loss": 1.9696, + "step": 259 + }, + { + "epoch": 0.025390625, + "grad_norm": 0.366953581571579, + "learning_rate": 0.00043333333333333337, + "loss": 2.031, + "step": 260 + }, + { + "epoch": 0.02548828125, + "grad_norm": 0.34289368987083435, + "learning_rate": 0.000435, + "loss": 1.9968, + "step": 261 + }, + { + "epoch": 0.0255859375, + "grad_norm": 0.35170793533325195, + "learning_rate": 0.00043666666666666664, + "loss": 1.9963, + "step": 262 + }, + { + "epoch": 0.02568359375, + "grad_norm": 0.28625521063804626, + "learning_rate": 0.0004383333333333334, + "loss": 1.9932, + "step": 263 + }, + { + "epoch": 0.02578125, + "grad_norm": 0.2861610949039459, + "learning_rate": 0.00044, + "loss": 2.0297, + "step": 264 + }, + { + "epoch": 0.02587890625, + "grad_norm": 0.30467647314071655, + "learning_rate": 0.00044166666666666665, + "loss": 1.992, + "step": 265 + }, + { + "epoch": 0.0259765625, + "grad_norm": 0.31711357831954956, + "learning_rate": 0.00044333333333333334, + "loss": 2.0135, + "step": 266 + }, + { + "epoch": 0.02607421875, + "grad_norm": 0.341530978679657, + "learning_rate": 0.00044500000000000003, + "loss": 1.983, + "step": 267 + }, + { + "epoch": 0.026171875, + "grad_norm": 0.373901903629303, + "learning_rate": 0.00044666666666666666, + "loss": 2.0048, + "step": 268 + }, + { + "epoch": 0.02626953125, + "grad_norm": 0.3105134665966034, + "learning_rate": 0.0004483333333333333, + "loss": 2.0368, + "step": 269 + }, + { + "epoch": 0.0263671875, + "grad_norm": 0.29363134503364563, + "learning_rate": 0.00045000000000000004, + "loss": 1.978, + "step": 270 + }, + { + "epoch": 0.02646484375, + "grad_norm": 0.3060167133808136, + "learning_rate": 0.0004516666666666667, + "loss": 1.9479, + "step": 271 + }, + { + "epoch": 0.0265625, + "grad_norm": 0.30803290009498596, + "learning_rate": 0.0004533333333333333, + "loss": 1.9662, + "step": 272 + }, + { + "epoch": 0.02666015625, + "grad_norm": 0.3324045240879059, + "learning_rate": 0.000455, + "loss": 2.0299, + "step": 273 + }, + { + "epoch": 0.0267578125, + "grad_norm": 0.39051148295402527, + "learning_rate": 0.0004566666666666667, + "loss": 1.9856, + "step": 274 + }, + { + "epoch": 0.02685546875, + "grad_norm": 0.4288715124130249, + "learning_rate": 0.0004583333333333333, + "loss": 2.0264, + "step": 275 + }, + { + "epoch": 0.026953125, + "grad_norm": 0.34478962421417236, + "learning_rate": 0.00046, + "loss": 1.9824, + "step": 276 + }, + { + "epoch": 0.02705078125, + "grad_norm": 0.2766290009021759, + "learning_rate": 0.0004616666666666667, + "loss": 2.0066, + "step": 277 + }, + { + "epoch": 0.0271484375, + "grad_norm": 0.2508682608604431, + "learning_rate": 0.00046333333333333334, + "loss": 1.9663, + "step": 278 + }, + { + "epoch": 0.02724609375, + "grad_norm": 0.26924827694892883, + "learning_rate": 0.000465, + "loss": 1.9903, + "step": 279 + }, + { + "epoch": 0.02734375, + "grad_norm": 0.27668496966362, + "learning_rate": 0.00046666666666666666, + "loss": 2.0097, + "step": 280 + }, + { + "epoch": 0.02744140625, + "grad_norm": 0.25026220083236694, + "learning_rate": 0.00046833333333333335, + "loss": 2.0583, + "step": 281 + }, + { + "epoch": 0.0275390625, + "grad_norm": 0.2158055454492569, + "learning_rate": 0.00047, + "loss": 2.0137, + "step": 282 + }, + { + "epoch": 0.02763671875, + "grad_norm": 0.22540244460105896, + "learning_rate": 0.0004716666666666667, + "loss": 1.994, + "step": 283 + }, + { + "epoch": 0.027734375, + "grad_norm": 0.26405519247055054, + "learning_rate": 0.00047333333333333336, + "loss": 2.0221, + "step": 284 + }, + { + "epoch": 0.02783203125, + "grad_norm": 0.2979099452495575, + "learning_rate": 0.000475, + "loss": 2.0047, + "step": 285 + }, + { + "epoch": 0.0279296875, + "grad_norm": 0.34131935238838196, + "learning_rate": 0.0004766666666666667, + "loss": 1.9907, + "step": 286 + }, + { + "epoch": 0.02802734375, + "grad_norm": 0.37178686261177063, + "learning_rate": 0.0004783333333333333, + "loss": 1.9806, + "step": 287 + }, + { + "epoch": 0.028125, + "grad_norm": 0.36835598945617676, + "learning_rate": 0.00048, + "loss": 2.0134, + "step": 288 + }, + { + "epoch": 0.02822265625, + "grad_norm": 0.29690125584602356, + "learning_rate": 0.0004816666666666667, + "loss": 2.0261, + "step": 289 + }, + { + "epoch": 0.0283203125, + "grad_norm": 0.2690771818161011, + "learning_rate": 0.00048333333333333334, + "loss": 1.9718, + "step": 290 + }, + { + "epoch": 0.02841796875, + "grad_norm": 0.3377201557159424, + "learning_rate": 0.00048499999999999997, + "loss": 1.99, + "step": 291 + }, + { + "epoch": 0.028515625, + "grad_norm": 0.34973010420799255, + "learning_rate": 0.0004866666666666667, + "loss": 1.9721, + "step": 292 + }, + { + "epoch": 0.02861328125, + "grad_norm": 0.3172457218170166, + "learning_rate": 0.0004883333333333333, + "loss": 1.9928, + "step": 293 + }, + { + "epoch": 0.0287109375, + "grad_norm": 0.34357598423957825, + "learning_rate": 0.00049, + "loss": 1.9995, + "step": 294 + }, + { + "epoch": 0.02880859375, + "grad_norm": 0.3824540376663208, + "learning_rate": 0.0004916666666666666, + "loss": 1.9772, + "step": 295 + }, + { + "epoch": 0.02890625, + "grad_norm": 0.3704535663127899, + "learning_rate": 0.0004933333333333334, + "loss": 1.9829, + "step": 296 + }, + { + "epoch": 0.02900390625, + "grad_norm": 0.2571757733821869, + "learning_rate": 0.000495, + "loss": 1.9966, + "step": 297 + }, + { + "epoch": 0.0291015625, + "grad_norm": 0.3970927894115448, + "learning_rate": 0.0004966666666666666, + "loss": 1.9841, + "step": 298 + }, + { + "epoch": 0.02919921875, + "grad_norm": 0.3420144319534302, + "learning_rate": 0.0004983333333333334, + "loss": 1.9747, + "step": 299 + }, + { + "epoch": 0.029296875, + "grad_norm": 0.30147823691368103, + "learning_rate": 0.0005, + "loss": 1.9999, + "step": 300 + }, + { + "epoch": 0.02939453125, + "grad_norm": 0.33727970719337463, + "learning_rate": 0.0004999999887622467, + "loss": 2.0084, + "step": 301 + }, + { + "epoch": 0.0294921875, + "grad_norm": 0.29407384991645813, + "learning_rate": 0.0004999999550489878, + "loss": 2.0144, + "step": 302 + }, + { + "epoch": 0.02958984375, + "grad_norm": 0.3489755094051361, + "learning_rate": 0.0004999998988602267, + "loss": 2.0058, + "step": 303 + }, + { + "epoch": 0.0296875, + "grad_norm": 0.3327770233154297, + "learning_rate": 0.0004999998201959691, + "loss": 2.0166, + "step": 304 + }, + { + "epoch": 0.02978515625, + "grad_norm": 0.2923370599746704, + "learning_rate": 0.0004999997190562227, + "loss": 2.0148, + "step": 305 + }, + { + "epoch": 0.0298828125, + "grad_norm": 0.31616437435150146, + "learning_rate": 0.0004999995954409976, + "loss": 1.9772, + "step": 306 + }, + { + "epoch": 0.02998046875, + "grad_norm": 0.22982288897037506, + "learning_rate": 0.0004999994493503064, + "loss": 2.0584, + "step": 307 + }, + { + "epoch": 0.030078125, + "grad_norm": 0.2886744737625122, + "learning_rate": 0.0004999992807841634, + "loss": 2.0114, + "step": 308 + }, + { + "epoch": 0.03017578125, + "grad_norm": 0.3027271032333374, + "learning_rate": 0.0004999990897425856, + "loss": 2.011, + "step": 309 + }, + { + "epoch": 0.0302734375, + "grad_norm": 0.3191162645816803, + "learning_rate": 0.0004999988762255922, + "loss": 1.9962, + "step": 310 + }, + { + "epoch": 0.03037109375, + "grad_norm": 0.34986981749534607, + "learning_rate": 0.0004999986402332042, + "loss": 1.9612, + "step": 311 + }, + { + "epoch": 0.03046875, + "grad_norm": 0.36431390047073364, + "learning_rate": 0.0004999983817654454, + "loss": 2.045, + "step": 312 + }, + { + "epoch": 0.03056640625, + "grad_norm": 0.4198042154312134, + "learning_rate": 0.0004999981008223416, + "loss": 2.0132, + "step": 313 + }, + { + "epoch": 0.0306640625, + "grad_norm": 0.43374890089035034, + "learning_rate": 0.0004999977974039207, + "loss": 2.0578, + "step": 314 + }, + { + "epoch": 0.03076171875, + "grad_norm": 0.3654812276363373, + "learning_rate": 0.0004999974715102132, + "loss": 1.9721, + "step": 315 + }, + { + "epoch": 0.030859375, + "grad_norm": 0.29420921206474304, + "learning_rate": 0.0004999971231412517, + "loss": 2.0296, + "step": 316 + }, + { + "epoch": 0.03095703125, + "grad_norm": 0.31979072093963623, + "learning_rate": 0.0004999967522970708, + "loss": 1.9623, + "step": 317 + }, + { + "epoch": 0.0310546875, + "grad_norm": 0.3570129871368408, + "learning_rate": 0.0004999963589777076, + "loss": 1.9966, + "step": 318 + }, + { + "epoch": 0.03115234375, + "grad_norm": 0.29743143916130066, + "learning_rate": 0.0004999959431832016, + "loss": 1.9535, + "step": 319 + }, + { + "epoch": 0.03125, + "grad_norm": 0.23641493916511536, + "learning_rate": 0.000499995504913594, + "loss": 2.0304, + "step": 320 + }, + { + "epoch": 0.03134765625, + "grad_norm": 0.3556622564792633, + "learning_rate": 0.0004999950441689288, + "loss": 1.971, + "step": 321 + }, + { + "epoch": 0.0314453125, + "grad_norm": 0.323939710855484, + "learning_rate": 0.0004999945609492519, + "loss": 2.0153, + "step": 322 + }, + { + "epoch": 0.03154296875, + "grad_norm": 0.22797244787216187, + "learning_rate": 0.0004999940552546118, + "loss": 1.9807, + "step": 323 + }, + { + "epoch": 0.031640625, + "grad_norm": 0.2641647458076477, + "learning_rate": 0.0004999935270850587, + "loss": 1.9988, + "step": 324 + }, + { + "epoch": 0.03173828125, + "grad_norm": 0.25289344787597656, + "learning_rate": 0.0004999929764406455, + "loss": 1.9748, + "step": 325 + }, + { + "epoch": 0.0318359375, + "grad_norm": 0.2377796769142151, + "learning_rate": 0.0004999924033214274, + "loss": 1.9983, + "step": 326 + }, + { + "epoch": 0.03193359375, + "grad_norm": 0.2711915671825409, + "learning_rate": 0.0004999918077274612, + "loss": 1.9643, + "step": 327 + }, + { + "epoch": 0.03203125, + "grad_norm": 0.2866462767124176, + "learning_rate": 0.0004999911896588068, + "loss": 2.0196, + "step": 328 + }, + { + "epoch": 0.03212890625, + "grad_norm": 0.25075578689575195, + "learning_rate": 0.0004999905491155257, + "loss": 2.0426, + "step": 329 + }, + { + "epoch": 0.0322265625, + "grad_norm": 0.266648530960083, + "learning_rate": 0.000499989886097682, + "loss": 2.0318, + "step": 330 + }, + { + "epoch": 0.03232421875, + "grad_norm": 0.22494247555732727, + "learning_rate": 0.0004999892006053421, + "loss": 1.9839, + "step": 331 + }, + { + "epoch": 0.032421875, + "grad_norm": 0.3179854452610016, + "learning_rate": 0.0004999884926385741, + "loss": 1.9981, + "step": 332 + }, + { + "epoch": 0.03251953125, + "grad_norm": 0.2754990756511688, + "learning_rate": 0.000499987762197449, + "loss": 1.9879, + "step": 333 + }, + { + "epoch": 0.0326171875, + "grad_norm": 0.2669137418270111, + "learning_rate": 0.0004999870092820395, + "loss": 2.0083, + "step": 334 + }, + { + "epoch": 0.03271484375, + "grad_norm": 0.31376007199287415, + "learning_rate": 0.0004999862338924212, + "loss": 2.0166, + "step": 335 + }, + { + "epoch": 0.0328125, + "grad_norm": 0.32645899057388306, + "learning_rate": 0.0004999854360286712, + "loss": 2.0019, + "step": 336 + }, + { + "epoch": 0.03291015625, + "grad_norm": 0.3550071120262146, + "learning_rate": 0.0004999846156908692, + "loss": 1.9744, + "step": 337 + }, + { + "epoch": 0.0330078125, + "grad_norm": 0.30171769857406616, + "learning_rate": 0.0004999837728790975, + "loss": 2.0231, + "step": 338 + }, + { + "epoch": 0.03310546875, + "grad_norm": 0.23128142952919006, + "learning_rate": 0.00049998290759344, + "loss": 2.001, + "step": 339 + }, + { + "epoch": 0.033203125, + "grad_norm": 0.2433364987373352, + "learning_rate": 0.0004999820198339832, + "loss": 1.9945, + "step": 340 + }, + { + "epoch": 0.03330078125, + "grad_norm": 0.30562201142311096, + "learning_rate": 0.0004999811096008159, + "loss": 2.0228, + "step": 341 + }, + { + "epoch": 0.0333984375, + "grad_norm": 0.33020487427711487, + "learning_rate": 0.0004999801768940287, + "loss": 1.9691, + "step": 342 + }, + { + "epoch": 0.03349609375, + "grad_norm": 0.3125375807285309, + "learning_rate": 0.0004999792217137151, + "loss": 1.9728, + "step": 343 + }, + { + "epoch": 0.03359375, + "grad_norm": 0.26769348978996277, + "learning_rate": 0.0004999782440599702, + "loss": 2.0066, + "step": 344 + }, + { + "epoch": 0.03369140625, + "grad_norm": 0.3596431314945221, + "learning_rate": 0.0004999772439328921, + "loss": 2.0426, + "step": 345 + }, + { + "epoch": 0.0337890625, + "grad_norm": 0.34117281436920166, + "learning_rate": 0.0004999762213325803, + "loss": 1.9985, + "step": 346 + }, + { + "epoch": 0.03388671875, + "grad_norm": 0.2922564148902893, + "learning_rate": 0.0004999751762591371, + "loss": 2.0346, + "step": 347 + }, + { + "epoch": 0.033984375, + "grad_norm": 0.2813419997692108, + "learning_rate": 0.0004999741087126669, + "loss": 2.0573, + "step": 348 + }, + { + "epoch": 0.03408203125, + "grad_norm": 0.27346089482307434, + "learning_rate": 0.0004999730186932764, + "loss": 1.9655, + "step": 349 + }, + { + "epoch": 0.0341796875, + "grad_norm": 0.24704065918922424, + "learning_rate": 0.0004999719062010745, + "loss": 1.9542, + "step": 350 + }, + { + "epoch": 0.03427734375, + "grad_norm": 0.24761976301670074, + "learning_rate": 0.000499970771236172, + "loss": 1.9526, + "step": 351 + }, + { + "epoch": 0.034375, + "grad_norm": 0.2508860230445862, + "learning_rate": 0.0004999696137986826, + "loss": 2.0119, + "step": 352 + }, + { + "epoch": 0.03447265625, + "grad_norm": 0.24160990118980408, + "learning_rate": 0.0004999684338887219, + "loss": 2.0148, + "step": 353 + }, + { + "epoch": 0.0345703125, + "grad_norm": 0.2879098355770111, + "learning_rate": 0.0004999672315064076, + "loss": 2.0206, + "step": 354 + }, + { + "epoch": 0.03466796875, + "grad_norm": 0.28366580605506897, + "learning_rate": 0.0004999660066518601, + "loss": 2.0737, + "step": 355 + }, + { + "epoch": 0.034765625, + "grad_norm": 0.3401612937450409, + "learning_rate": 0.0004999647593252013, + "loss": 1.9868, + "step": 356 + }, + { + "epoch": 0.03486328125, + "grad_norm": 0.4696269631385803, + "learning_rate": 0.0004999634895265562, + "loss": 2.0079, + "step": 357 + }, + { + "epoch": 0.0349609375, + "grad_norm": 0.596168041229248, + "learning_rate": 0.0004999621972560515, + "loss": 2.0084, + "step": 358 + }, + { + "epoch": 0.03505859375, + "grad_norm": 0.5319205522537231, + "learning_rate": 0.0004999608825138162, + "loss": 2.0084, + "step": 359 + }, + { + "epoch": 0.03515625, + "grad_norm": 0.24298223853111267, + "learning_rate": 0.0004999595452999818, + "loss": 1.9183, + "step": 360 + }, + { + "epoch": 0.03525390625, + "grad_norm": 0.39650923013687134, + "learning_rate": 0.0004999581856146817, + "loss": 1.9886, + "step": 361 + }, + { + "epoch": 0.0353515625, + "grad_norm": 0.3819667100906372, + "learning_rate": 0.0004999568034580518, + "loss": 2.0047, + "step": 362 + }, + { + "epoch": 0.03544921875, + "grad_norm": 0.3298279345035553, + "learning_rate": 0.0004999553988302303, + "loss": 1.9466, + "step": 363 + }, + { + "epoch": 0.035546875, + "grad_norm": 0.2343115359544754, + "learning_rate": 0.0004999539717313573, + "loss": 1.9828, + "step": 364 + }, + { + "epoch": 0.03564453125, + "grad_norm": 0.31238943338394165, + "learning_rate": 0.0004999525221615755, + "loss": 2.0721, + "step": 365 + }, + { + "epoch": 0.0357421875, + "grad_norm": 0.27481910586357117, + "learning_rate": 0.0004999510501210295, + "loss": 2.0106, + "step": 366 + }, + { + "epoch": 0.03583984375, + "grad_norm": 0.24870915710926056, + "learning_rate": 0.0004999495556098666, + "loss": 1.9846, + "step": 367 + }, + { + "epoch": 0.0359375, + "grad_norm": 0.2666539251804352, + "learning_rate": 0.0004999480386282359, + "loss": 1.9988, + "step": 368 + }, + { + "epoch": 0.03603515625, + "grad_norm": 0.23231451213359833, + "learning_rate": 0.000499946499176289, + "loss": 2.0235, + "step": 369 + }, + { + "epoch": 0.0361328125, + "grad_norm": 0.19276577234268188, + "learning_rate": 0.0004999449372541798, + "loss": 1.9832, + "step": 370 + }, + { + "epoch": 0.03623046875, + "grad_norm": 0.23545822501182556, + "learning_rate": 0.000499943352862064, + "loss": 1.9879, + "step": 371 + }, + { + "epoch": 0.036328125, + "grad_norm": 0.2608807384967804, + "learning_rate": 0.0004999417460001002, + "loss": 2.047, + "step": 372 + }, + { + "epoch": 0.03642578125, + "grad_norm": 0.27395492792129517, + "learning_rate": 0.0004999401166684487, + "loss": 1.9496, + "step": 373 + }, + { + "epoch": 0.0365234375, + "grad_norm": 0.24661333858966827, + "learning_rate": 0.0004999384648672724, + "loss": 1.9924, + "step": 374 + }, + { + "epoch": 0.03662109375, + "grad_norm": 0.23559828102588654, + "learning_rate": 0.0004999367905967362, + "loss": 1.979, + "step": 375 + }, + { + "epoch": 0.03671875, + "grad_norm": 0.2556357681751251, + "learning_rate": 0.0004999350938570074, + "loss": 1.9793, + "step": 376 + }, + { + "epoch": 0.03681640625, + "grad_norm": 0.29081088304519653, + "learning_rate": 0.0004999333746482555, + "loss": 1.962, + "step": 377 + }, + { + "epoch": 0.0369140625, + "grad_norm": 0.2583800256252289, + "learning_rate": 0.0004999316329706521, + "loss": 1.9765, + "step": 378 + }, + { + "epoch": 0.03701171875, + "grad_norm": 0.2506580054759979, + "learning_rate": 0.0004999298688243714, + "loss": 2.0302, + "step": 379 + }, + { + "epoch": 0.037109375, + "grad_norm": 0.289530485868454, + "learning_rate": 0.0004999280822095895, + "loss": 1.9514, + "step": 380 + }, + { + "epoch": 0.03720703125, + "grad_norm": 0.24479152262210846, + "learning_rate": 0.0004999262731264848, + "loss": 1.9742, + "step": 381 + }, + { + "epoch": 0.0373046875, + "grad_norm": 0.23676550388336182, + "learning_rate": 0.0004999244415752381, + "loss": 2.0161, + "step": 382 + }, + { + "epoch": 0.03740234375, + "grad_norm": 0.2447502315044403, + "learning_rate": 0.0004999225875560323, + "loss": 2.0114, + "step": 383 + }, + { + "epoch": 0.0375, + "grad_norm": 0.20993874967098236, + "learning_rate": 0.0004999207110690528, + "loss": 1.9786, + "step": 384 + }, + { + "epoch": 0.03759765625, + "grad_norm": 0.23405557870864868, + "learning_rate": 0.0004999188121144867, + "loss": 2.0004, + "step": 385 + }, + { + "epoch": 0.0376953125, + "grad_norm": 0.2310025990009308, + "learning_rate": 0.0004999168906925238, + "loss": 1.924, + "step": 386 + }, + { + "epoch": 0.03779296875, + "grad_norm": 0.2507460117340088, + "learning_rate": 0.0004999149468033564, + "loss": 2.0301, + "step": 387 + }, + { + "epoch": 0.037890625, + "grad_norm": 0.30916762351989746, + "learning_rate": 0.0004999129804471782, + "loss": 1.9948, + "step": 388 + }, + { + "epoch": 0.03798828125, + "grad_norm": 0.42094507813453674, + "learning_rate": 0.0004999109916241858, + "loss": 2.0128, + "step": 389 + }, + { + "epoch": 0.0380859375, + "grad_norm": 0.5398088693618774, + "learning_rate": 0.0004999089803345779, + "loss": 1.9866, + "step": 390 + }, + { + "epoch": 0.03818359375, + "grad_norm": 0.4601620137691498, + "learning_rate": 0.0004999069465785554, + "loss": 1.9807, + "step": 391 + }, + { + "epoch": 0.03828125, + "grad_norm": 0.24873095750808716, + "learning_rate": 0.0004999048903563213, + "loss": 1.9225, + "step": 392 + }, + { + "epoch": 0.03837890625, + "grad_norm": 0.37363573908805847, + "learning_rate": 0.0004999028116680814, + "loss": 2.0212, + "step": 393 + }, + { + "epoch": 0.0384765625, + "grad_norm": 0.27399107813835144, + "learning_rate": 0.0004999007105140428, + "loss": 1.9797, + "step": 394 + }, + { + "epoch": 0.03857421875, + "grad_norm": 0.2717953622341156, + "learning_rate": 0.0004998985868944158, + "loss": 2.0119, + "step": 395 + }, + { + "epoch": 0.038671875, + "grad_norm": 0.2766648530960083, + "learning_rate": 0.0004998964408094124, + "loss": 1.9936, + "step": 396 + }, + { + "epoch": 0.03876953125, + "grad_norm": 0.2989843487739563, + "learning_rate": 0.0004998942722592469, + "loss": 2.0261, + "step": 397 + }, + { + "epoch": 0.0388671875, + "grad_norm": 0.31954097747802734, + "learning_rate": 0.000499892081244136, + "loss": 1.9713, + "step": 398 + }, + { + "epoch": 0.03896484375, + "grad_norm": 0.31199777126312256, + "learning_rate": 0.0004998898677642987, + "loss": 1.9548, + "step": 399 + }, + { + "epoch": 0.0390625, + "grad_norm": 0.2475418746471405, + "learning_rate": 0.0004998876318199557, + "loss": 1.9878, + "step": 400 + }, + { + "epoch": 0.03916015625, + "grad_norm": 0.25001809000968933, + "learning_rate": 0.0004998853734113308, + "loss": 1.991, + "step": 401 + }, + { + "epoch": 0.0392578125, + "grad_norm": 0.2468329817056656, + "learning_rate": 0.0004998830925386492, + "loss": 1.9988, + "step": 402 + }, + { + "epoch": 0.03935546875, + "grad_norm": 0.23273305594921112, + "learning_rate": 0.000499880789202139, + "loss": 2.0138, + "step": 403 + }, + { + "epoch": 0.039453125, + "grad_norm": 0.21512626111507416, + "learning_rate": 0.0004998784634020303, + "loss": 1.9326, + "step": 404 + }, + { + "epoch": 0.03955078125, + "grad_norm": 0.2123369723558426, + "learning_rate": 0.0004998761151385554, + "loss": 1.9954, + "step": 405 + }, + { + "epoch": 0.0396484375, + "grad_norm": 0.20985403656959534, + "learning_rate": 0.0004998737444119488, + "loss": 1.996, + "step": 406 + }, + { + "epoch": 0.03974609375, + "grad_norm": 0.19709332287311554, + "learning_rate": 0.0004998713512224473, + "loss": 1.9642, + "step": 407 + }, + { + "epoch": 0.03984375, + "grad_norm": 0.21997478604316711, + "learning_rate": 0.00049986893557029, + "loss": 1.9925, + "step": 408 + }, + { + "epoch": 0.03994140625, + "grad_norm": 0.2419801503419876, + "learning_rate": 0.0004998664974557182, + "loss": 2.0262, + "step": 409 + }, + { + "epoch": 0.0400390625, + "grad_norm": 0.2527197003364563, + "learning_rate": 0.0004998640368789754, + "loss": 2.0011, + "step": 410 + }, + { + "epoch": 0.04013671875, + "grad_norm": 0.2539777159690857, + "learning_rate": 0.0004998615538403074, + "loss": 1.9799, + "step": 411 + }, + { + "epoch": 0.040234375, + "grad_norm": 0.2419251948595047, + "learning_rate": 0.0004998590483399623, + "loss": 2.0035, + "step": 412 + }, + { + "epoch": 0.04033203125, + "grad_norm": 0.26659199595451355, + "learning_rate": 0.0004998565203781904, + "loss": 1.9743, + "step": 413 + }, + { + "epoch": 0.0404296875, + "grad_norm": 0.3216726779937744, + "learning_rate": 0.0004998539699552441, + "loss": 1.979, + "step": 414 + }, + { + "epoch": 0.04052734375, + "grad_norm": 0.2780512571334839, + "learning_rate": 0.0004998513970713783, + "loss": 1.9892, + "step": 415 + }, + { + "epoch": 0.040625, + "grad_norm": 0.2947809100151062, + "learning_rate": 0.00049984880172685, + "loss": 1.9836, + "step": 416 + }, + { + "epoch": 0.04072265625, + "grad_norm": 0.33937209844589233, + "learning_rate": 0.0004998461839219182, + "loss": 2.0007, + "step": 417 + }, + { + "epoch": 0.0408203125, + "grad_norm": 0.3530365228652954, + "learning_rate": 0.0004998435436568446, + "loss": 1.9462, + "step": 418 + }, + { + "epoch": 0.04091796875, + "grad_norm": 0.34433260560035706, + "learning_rate": 0.000499840880931893, + "loss": 1.9733, + "step": 419 + }, + { + "epoch": 0.041015625, + "grad_norm": 0.323081910610199, + "learning_rate": 0.0004998381957473293, + "loss": 1.9522, + "step": 420 + }, + { + "epoch": 0.04111328125, + "grad_norm": 0.22895868122577667, + "learning_rate": 0.0004998354881034217, + "loss": 1.9672, + "step": 421 + }, + { + "epoch": 0.0412109375, + "grad_norm": 0.2543105185031891, + "learning_rate": 0.0004998327580004408, + "loss": 1.9429, + "step": 422 + }, + { + "epoch": 0.04130859375, + "grad_norm": 0.2795581519603729, + "learning_rate": 0.0004998300054386591, + "loss": 1.9902, + "step": 423 + }, + { + "epoch": 0.04140625, + "grad_norm": 0.2201048880815506, + "learning_rate": 0.0004998272304183517, + "loss": 1.9365, + "step": 424 + }, + { + "epoch": 0.04150390625, + "grad_norm": 0.26653197407722473, + "learning_rate": 0.0004998244329397958, + "loss": 2.0178, + "step": 425 + }, + { + "epoch": 0.0416015625, + "grad_norm": 0.26959821581840515, + "learning_rate": 0.0004998216130032708, + "loss": 1.9625, + "step": 426 + }, + { + "epoch": 0.04169921875, + "grad_norm": 0.2673629820346832, + "learning_rate": 0.0004998187706090584, + "loss": 1.995, + "step": 427 + }, + { + "epoch": 0.041796875, + "grad_norm": 0.3190925121307373, + "learning_rate": 0.0004998159057574426, + "loss": 2.0094, + "step": 428 + }, + { + "epoch": 0.04189453125, + "grad_norm": 0.3376927673816681, + "learning_rate": 0.0004998130184487094, + "loss": 2.0535, + "step": 429 + }, + { + "epoch": 0.0419921875, + "grad_norm": 0.35333067178726196, + "learning_rate": 0.0004998101086831474, + "loss": 2.0256, + "step": 430 + }, + { + "epoch": 0.04208984375, + "grad_norm": 0.3051585853099823, + "learning_rate": 0.0004998071764610471, + "loss": 1.9798, + "step": 431 + }, + { + "epoch": 0.0421875, + "grad_norm": 0.21881523728370667, + "learning_rate": 0.0004998042217827015, + "loss": 1.9726, + "step": 432 + }, + { + "epoch": 0.04228515625, + "grad_norm": 0.2296576052904129, + "learning_rate": 0.0004998012446484057, + "loss": 1.9675, + "step": 433 + }, + { + "epoch": 0.0423828125, + "grad_norm": 0.24055050313472748, + "learning_rate": 0.0004997982450584572, + "loss": 1.9488, + "step": 434 + }, + { + "epoch": 0.04248046875, + "grad_norm": 0.22267191112041473, + "learning_rate": 0.0004997952230131555, + "loss": 1.9617, + "step": 435 + }, + { + "epoch": 0.042578125, + "grad_norm": 0.2590855658054352, + "learning_rate": 0.0004997921785128026, + "loss": 1.9745, + "step": 436 + }, + { + "epoch": 0.04267578125, + "grad_norm": 0.2995806336402893, + "learning_rate": 0.0004997891115577025, + "loss": 1.9667, + "step": 437 + }, + { + "epoch": 0.0427734375, + "grad_norm": 0.3083277642726898, + "learning_rate": 0.0004997860221481616, + "loss": 1.9661, + "step": 438 + }, + { + "epoch": 0.04287109375, + "grad_norm": 0.31743094325065613, + "learning_rate": 0.0004997829102844885, + "loss": 2.0172, + "step": 439 + }, + { + "epoch": 0.04296875, + "grad_norm": 0.37507855892181396, + "learning_rate": 0.0004997797759669941, + "loss": 2.0211, + "step": 440 + }, + { + "epoch": 0.04306640625, + "grad_norm": 0.35785287618637085, + "learning_rate": 0.0004997766191959914, + "loss": 2.038, + "step": 441 + }, + { + "epoch": 0.0431640625, + "grad_norm": 0.3099903166294098, + "learning_rate": 0.0004997734399717958, + "loss": 1.9879, + "step": 442 + }, + { + "epoch": 0.04326171875, + "grad_norm": 0.32211577892303467, + "learning_rate": 0.0004997702382947248, + "loss": 1.987, + "step": 443 + }, + { + "epoch": 0.043359375, + "grad_norm": 0.24999700486660004, + "learning_rate": 0.0004997670141650984, + "loss": 1.9765, + "step": 444 + }, + { + "epoch": 0.04345703125, + "grad_norm": 0.23301391303539276, + "learning_rate": 0.0004997637675832386, + "loss": 1.9592, + "step": 445 + }, + { + "epoch": 0.0435546875, + "grad_norm": 0.22780932486057281, + "learning_rate": 0.0004997604985494694, + "loss": 1.9762, + "step": 446 + }, + { + "epoch": 0.04365234375, + "grad_norm": 0.2347191423177719, + "learning_rate": 0.0004997572070641178, + "loss": 1.9791, + "step": 447 + }, + { + "epoch": 0.04375, + "grad_norm": 0.3288608193397522, + "learning_rate": 0.0004997538931275123, + "loss": 1.9664, + "step": 448 + }, + { + "epoch": 0.04384765625, + "grad_norm": 0.29897212982177734, + "learning_rate": 0.000499750556739984, + "loss": 1.9876, + "step": 449 + }, + { + "epoch": 0.0439453125, + "grad_norm": 0.24389693140983582, + "learning_rate": 0.0004997471979018663, + "loss": 1.9742, + "step": 450 + }, + { + "epoch": 0.04404296875, + "grad_norm": 0.3172384798526764, + "learning_rate": 0.0004997438166134945, + "loss": 1.969, + "step": 451 + }, + { + "epoch": 0.044140625, + "grad_norm": 0.22951611876487732, + "learning_rate": 0.0004997404128752065, + "loss": 1.9832, + "step": 452 + }, + { + "epoch": 0.04423828125, + "grad_norm": 0.2356499582529068, + "learning_rate": 0.0004997369866873423, + "loss": 2.0236, + "step": 453 + }, + { + "epoch": 0.0443359375, + "grad_norm": 0.2687929570674896, + "learning_rate": 0.000499733538050244, + "loss": 1.9521, + "step": 454 + }, + { + "epoch": 0.04443359375, + "grad_norm": 0.2512679398059845, + "learning_rate": 0.0004997300669642564, + "loss": 1.9754, + "step": 455 + }, + { + "epoch": 0.04453125, + "grad_norm": 0.22110989689826965, + "learning_rate": 0.0004997265734297259, + "loss": 1.9931, + "step": 456 + }, + { + "epoch": 0.04462890625, + "grad_norm": 0.2548464834690094, + "learning_rate": 0.0004997230574470017, + "loss": 2.001, + "step": 457 + }, + { + "epoch": 0.0447265625, + "grad_norm": 0.3107970356941223, + "learning_rate": 0.0004997195190164349, + "loss": 1.9855, + "step": 458 + }, + { + "epoch": 0.04482421875, + "grad_norm": 0.2989678680896759, + "learning_rate": 0.0004997159581383789, + "loss": 1.9646, + "step": 459 + }, + { + "epoch": 0.044921875, + "grad_norm": 0.3651764690876007, + "learning_rate": 0.0004997123748131896, + "loss": 1.9865, + "step": 460 + }, + { + "epoch": 0.04501953125, + "grad_norm": 0.41693389415740967, + "learning_rate": 0.0004997087690412248, + "loss": 1.9756, + "step": 461 + }, + { + "epoch": 0.0451171875, + "grad_norm": 0.425959974527359, + "learning_rate": 0.0004997051408228447, + "loss": 1.9887, + "step": 462 + }, + { + "epoch": 0.04521484375, + "grad_norm": 0.38966843485832214, + "learning_rate": 0.0004997014901584118, + "loss": 1.9532, + "step": 463 + }, + { + "epoch": 0.0453125, + "grad_norm": 0.29931432008743286, + "learning_rate": 0.0004996978170482906, + "loss": 1.9806, + "step": 464 + }, + { + "epoch": 0.04541015625, + "grad_norm": 0.23077397048473358, + "learning_rate": 0.0004996941214928482, + "loss": 2.0258, + "step": 465 + }, + { + "epoch": 0.0455078125, + "grad_norm": 0.32101985812187195, + "learning_rate": 0.0004996904034924536, + "loss": 2.0023, + "step": 466 + }, + { + "epoch": 0.04560546875, + "grad_norm": 0.3225870430469513, + "learning_rate": 0.0004996866630474783, + "loss": 1.9385, + "step": 467 + }, + { + "epoch": 0.045703125, + "grad_norm": 0.2592551112174988, + "learning_rate": 0.0004996829001582959, + "loss": 2.0176, + "step": 468 + }, + { + "epoch": 0.04580078125, + "grad_norm": 0.24367393553256989, + "learning_rate": 0.0004996791148252824, + "loss": 1.9844, + "step": 469 + }, + { + "epoch": 0.0458984375, + "grad_norm": 0.26423218846321106, + "learning_rate": 0.0004996753070488156, + "loss": 1.9675, + "step": 470 + }, + { + "epoch": 0.04599609375, + "grad_norm": 0.27801692485809326, + "learning_rate": 0.0004996714768292762, + "loss": 1.9265, + "step": 471 + }, + { + "epoch": 0.04609375, + "grad_norm": 0.3210003077983856, + "learning_rate": 0.0004996676241670467, + "loss": 1.9937, + "step": 472 + }, + { + "epoch": 0.04619140625, + "grad_norm": 0.3082759380340576, + "learning_rate": 0.0004996637490625118, + "loss": 1.9997, + "step": 473 + }, + { + "epoch": 0.0462890625, + "grad_norm": 0.2486785650253296, + "learning_rate": 0.0004996598515160589, + "loss": 2.0687, + "step": 474 + }, + { + "epoch": 0.04638671875, + "grad_norm": 0.31927499175071716, + "learning_rate": 0.000499655931528077, + "loss": 1.9853, + "step": 475 + }, + { + "epoch": 0.046484375, + "grad_norm": 0.2736678123474121, + "learning_rate": 0.0004996519890989578, + "loss": 1.9916, + "step": 476 + }, + { + "epoch": 0.04658203125, + "grad_norm": 0.29522353410720825, + "learning_rate": 0.0004996480242290952, + "loss": 1.9554, + "step": 477 + }, + { + "epoch": 0.0466796875, + "grad_norm": 0.26841938495635986, + "learning_rate": 0.0004996440369188851, + "loss": 1.9275, + "step": 478 + }, + { + "epoch": 0.04677734375, + "grad_norm": 0.2695488929748535, + "learning_rate": 0.0004996400271687259, + "loss": 1.9898, + "step": 479 + }, + { + "epoch": 0.046875, + "grad_norm": 0.36634454131126404, + "learning_rate": 0.0004996359949790181, + "loss": 1.9799, + "step": 480 + }, + { + "epoch": 0.04697265625, + "grad_norm": 0.37612974643707275, + "learning_rate": 0.0004996319403501647, + "loss": 1.9645, + "step": 481 + }, + { + "epoch": 0.0470703125, + "grad_norm": 0.29007261991500854, + "learning_rate": 0.0004996278632825703, + "loss": 1.9486, + "step": 482 + }, + { + "epoch": 0.04716796875, + "grad_norm": 0.30373314023017883, + "learning_rate": 0.0004996237637766424, + "loss": 1.9913, + "step": 483 + }, + { + "epoch": 0.047265625, + "grad_norm": 0.2949075400829315, + "learning_rate": 0.0004996196418327906, + "loss": 1.9894, + "step": 484 + }, + { + "epoch": 0.04736328125, + "grad_norm": 0.30601680278778076, + "learning_rate": 0.0004996154974514264, + "loss": 1.9835, + "step": 485 + }, + { + "epoch": 0.0474609375, + "grad_norm": 0.2896345853805542, + "learning_rate": 0.000499611330632964, + "loss": 2.0043, + "step": 486 + }, + { + "epoch": 0.04755859375, + "grad_norm": 0.3836851418018341, + "learning_rate": 0.0004996071413778195, + "loss": 1.9782, + "step": 487 + }, + { + "epoch": 0.04765625, + "grad_norm": 0.4167245626449585, + "learning_rate": 0.0004996029296864114, + "loss": 1.9792, + "step": 488 + }, + { + "epoch": 0.04775390625, + "grad_norm": 0.3247157335281372, + "learning_rate": 0.0004995986955591606, + "loss": 1.9548, + "step": 489 + }, + { + "epoch": 0.0478515625, + "grad_norm": 0.36726462841033936, + "learning_rate": 0.0004995944389964897, + "loss": 1.9844, + "step": 490 + }, + { + "epoch": 0.04794921875, + "grad_norm": 0.27440258860588074, + "learning_rate": 0.0004995901599988241, + "loss": 1.9608, + "step": 491 + }, + { + "epoch": 0.048046875, + "grad_norm": 0.271771103143692, + "learning_rate": 0.0004995858585665912, + "loss": 2.022, + "step": 492 + }, + { + "epoch": 0.04814453125, + "grad_norm": 0.3079121708869934, + "learning_rate": 0.0004995815347002208, + "loss": 1.9804, + "step": 493 + }, + { + "epoch": 0.0482421875, + "grad_norm": 0.2976233661174774, + "learning_rate": 0.0004995771884001445, + "loss": 1.9679, + "step": 494 + }, + { + "epoch": 0.04833984375, + "grad_norm": 0.31318438053131104, + "learning_rate": 0.0004995728196667969, + "loss": 2.0342, + "step": 495 + }, + { + "epoch": 0.0484375, + "grad_norm": 0.2397848516702652, + "learning_rate": 0.0004995684285006139, + "loss": 1.9801, + "step": 496 + }, + { + "epoch": 0.04853515625, + "grad_norm": 0.2205589860677719, + "learning_rate": 0.0004995640149020346, + "loss": 2.0088, + "step": 497 + }, + { + "epoch": 0.0486328125, + "grad_norm": 0.23892101645469666, + "learning_rate": 0.0004995595788714995, + "loss": 1.9888, + "step": 498 + }, + { + "epoch": 0.04873046875, + "grad_norm": 0.2999497354030609, + "learning_rate": 0.000499555120409452, + "loss": 1.9896, + "step": 499 + }, + { + "epoch": 0.048828125, + "grad_norm": 0.3309103548526764, + "learning_rate": 0.0004995506395163372, + "loss": 1.9394, + "step": 500 + }, + { + "epoch": 0.04892578125, + "grad_norm": 0.36243122816085815, + "learning_rate": 0.000499546136192603, + "loss": 1.9604, + "step": 501 + }, + { + "epoch": 0.0490234375, + "grad_norm": 0.3052852153778076, + "learning_rate": 0.0004995416104386991, + "loss": 1.9887, + "step": 502 + }, + { + "epoch": 0.04912109375, + "grad_norm": 0.2957259714603424, + "learning_rate": 0.0004995370622550775, + "loss": 1.9876, + "step": 503 + }, + { + "epoch": 0.04921875, + "grad_norm": 0.3991003632545471, + "learning_rate": 0.0004995324916421926, + "loss": 1.9932, + "step": 504 + }, + { + "epoch": 0.04931640625, + "grad_norm": 0.27617356181144714, + "learning_rate": 0.000499527898600501, + "loss": 1.9387, + "step": 505 + }, + { + "epoch": 0.0494140625, + "grad_norm": 0.2820669412612915, + "learning_rate": 0.0004995232831304614, + "loss": 2.012, + "step": 506 + }, + { + "epoch": 0.04951171875, + "grad_norm": 0.3026665449142456, + "learning_rate": 0.0004995186452325351, + "loss": 1.9709, + "step": 507 + }, + { + "epoch": 0.049609375, + "grad_norm": 0.2331731766462326, + "learning_rate": 0.000499513984907185, + "loss": 1.9617, + "step": 508 + }, + { + "epoch": 0.04970703125, + "grad_norm": 0.2791639268398285, + "learning_rate": 0.0004995093021548768, + "loss": 1.9635, + "step": 509 + }, + { + "epoch": 0.0498046875, + "grad_norm": 0.257519006729126, + "learning_rate": 0.0004995045969760785, + "loss": 1.8908, + "step": 510 + }, + { + "epoch": 0.04990234375, + "grad_norm": 0.26278433203697205, + "learning_rate": 0.0004994998693712598, + "loss": 1.9418, + "step": 511 + }, + { + "epoch": 0.05, + "grad_norm": 0.2534024715423584, + "learning_rate": 0.0004994951193408929, + "loss": 1.9821, + "step": 512 + }, + { + "epoch": 0.05009765625, + "grad_norm": 0.272348552942276, + "learning_rate": 0.0004994903468854527, + "loss": 1.982, + "step": 513 + }, + { + "epoch": 0.0501953125, + "grad_norm": 0.23883765935897827, + "learning_rate": 0.0004994855520054154, + "loss": 2.006, + "step": 514 + }, + { + "epoch": 0.05029296875, + "grad_norm": 0.25627830624580383, + "learning_rate": 0.0004994807347012603, + "loss": 2.0008, + "step": 515 + }, + { + "epoch": 0.050390625, + "grad_norm": 0.33997049927711487, + "learning_rate": 0.0004994758949734686, + "loss": 2.0339, + "step": 516 + }, + { + "epoch": 0.05048828125, + "grad_norm": 0.4115971028804779, + "learning_rate": 0.0004994710328225236, + "loss": 1.9982, + "step": 517 + }, + { + "epoch": 0.0505859375, + "grad_norm": 0.4417625069618225, + "learning_rate": 0.000499466148248911, + "loss": 1.9742, + "step": 518 + }, + { + "epoch": 0.05068359375, + "grad_norm": 0.4123833477497101, + "learning_rate": 0.0004994612412531189, + "loss": 2.0148, + "step": 519 + }, + { + "epoch": 0.05078125, + "grad_norm": 0.2234133630990982, + "learning_rate": 0.0004994563118356373, + "loss": 1.9679, + "step": 520 + }, + { + "epoch": 0.05087890625, + "grad_norm": 0.35439473390579224, + "learning_rate": 0.0004994513599969586, + "loss": 1.9602, + "step": 521 + }, + { + "epoch": 0.0509765625, + "grad_norm": 0.32346996665000916, + "learning_rate": 0.0004994463857375776, + "loss": 2.0106, + "step": 522 + }, + { + "epoch": 0.05107421875, + "grad_norm": 0.24506594240665436, + "learning_rate": 0.000499441389057991, + "loss": 1.993, + "step": 523 + }, + { + "epoch": 0.051171875, + "grad_norm": 0.26808494329452515, + "learning_rate": 0.000499436369958698, + "loss": 1.9764, + "step": 524 + }, + { + "epoch": 0.05126953125, + "grad_norm": 0.21936193108558655, + "learning_rate": 0.0004994313284401999, + "loss": 1.9619, + "step": 525 + }, + { + "epoch": 0.0513671875, + "grad_norm": 0.2545083165168762, + "learning_rate": 0.0004994262645030005, + "loss": 1.9825, + "step": 526 + }, + { + "epoch": 0.05146484375, + "grad_norm": 0.26330727338790894, + "learning_rate": 0.0004994211781476055, + "loss": 1.9512, + "step": 527 + }, + { + "epoch": 0.0515625, + "grad_norm": 0.2708059549331665, + "learning_rate": 0.0004994160693745229, + "loss": 1.9761, + "step": 528 + }, + { + "epoch": 0.05166015625, + "grad_norm": 0.2882954478263855, + "learning_rate": 0.0004994109381842632, + "loss": 2.0223, + "step": 529 + }, + { + "epoch": 0.0517578125, + "grad_norm": 0.25243133306503296, + "learning_rate": 0.0004994057845773389, + "loss": 1.976, + "step": 530 + }, + { + "epoch": 0.05185546875, + "grad_norm": 0.23666390776634216, + "learning_rate": 0.0004994006085542648, + "loss": 1.9729, + "step": 531 + }, + { + "epoch": 0.051953125, + "grad_norm": 0.24242763221263885, + "learning_rate": 0.0004993954101155578, + "loss": 1.9452, + "step": 532 + }, + { + "epoch": 0.05205078125, + "grad_norm": 0.259750634431839, + "learning_rate": 0.0004993901892617373, + "loss": 1.9582, + "step": 533 + }, + { + "epoch": 0.0521484375, + "grad_norm": 0.28516751527786255, + "learning_rate": 0.0004993849459933249, + "loss": 1.9367, + "step": 534 + }, + { + "epoch": 0.05224609375, + "grad_norm": 0.26534441113471985, + "learning_rate": 0.0004993796803108442, + "loss": 2.0137, + "step": 535 + }, + { + "epoch": 0.05234375, + "grad_norm": 0.28771716356277466, + "learning_rate": 0.0004993743922148213, + "loss": 1.927, + "step": 536 + }, + { + "epoch": 0.05244140625, + "grad_norm": 0.2747785449028015, + "learning_rate": 0.0004993690817057844, + "loss": 1.9729, + "step": 537 + }, + { + "epoch": 0.0525390625, + "grad_norm": 0.29758942127227783, + "learning_rate": 0.0004993637487842639, + "loss": 1.9713, + "step": 538 + }, + { + "epoch": 0.05263671875, + "grad_norm": 0.3059535026550293, + "learning_rate": 0.0004993583934507927, + "loss": 2.0313, + "step": 539 + }, + { + "epoch": 0.052734375, + "grad_norm": 0.36820656061172485, + "learning_rate": 0.0004993530157059056, + "loss": 1.9476, + "step": 540 + }, + { + "epoch": 0.05283203125, + "grad_norm": 0.39971432089805603, + "learning_rate": 0.0004993476155501396, + "loss": 2.0238, + "step": 541 + }, + { + "epoch": 0.0529296875, + "grad_norm": 0.3369678854942322, + "learning_rate": 0.0004993421929840346, + "loss": 1.9502, + "step": 542 + }, + { + "epoch": 0.05302734375, + "grad_norm": 0.3733840882778168, + "learning_rate": 0.000499336748008132, + "loss": 2.0114, + "step": 543 + }, + { + "epoch": 0.053125, + "grad_norm": 0.3251579999923706, + "learning_rate": 0.0004993312806229757, + "loss": 1.964, + "step": 544 + }, + { + "epoch": 0.05322265625, + "grad_norm": 0.4093327522277832, + "learning_rate": 0.0004993257908291117, + "loss": 1.9853, + "step": 545 + }, + { + "epoch": 0.0533203125, + "grad_norm": 0.26201343536376953, + "learning_rate": 0.0004993202786270888, + "loss": 2.0297, + "step": 546 + }, + { + "epoch": 0.05341796875, + "grad_norm": 0.27842357754707336, + "learning_rate": 0.0004993147440174572, + "loss": 2.003, + "step": 547 + }, + { + "epoch": 0.053515625, + "grad_norm": 0.2930687367916107, + "learning_rate": 0.00049930918700077, + "loss": 1.9634, + "step": 548 + }, + { + "epoch": 0.05361328125, + "grad_norm": 0.2799973785877228, + "learning_rate": 0.0004993036075775821, + "loss": 2.0023, + "step": 549 + }, + { + "epoch": 0.0537109375, + "grad_norm": 0.35084354877471924, + "learning_rate": 0.000499298005748451, + "loss": 2.0508, + "step": 550 + }, + { + "epoch": 0.05380859375, + "grad_norm": 0.2895161211490631, + "learning_rate": 0.0004992923815139362, + "loss": 2.0074, + "step": 551 + }, + { + "epoch": 0.05390625, + "grad_norm": 0.2637045979499817, + "learning_rate": 0.0004992867348745997, + "loss": 1.9559, + "step": 552 + }, + { + "epoch": 0.05400390625, + "grad_norm": 0.2837629020214081, + "learning_rate": 0.0004992810658310052, + "loss": 1.9603, + "step": 553 + }, + { + "epoch": 0.0541015625, + "grad_norm": 0.2644243836402893, + "learning_rate": 0.0004992753743837193, + "loss": 1.9767, + "step": 554 + }, + { + "epoch": 0.05419921875, + "grad_norm": 0.21815134584903717, + "learning_rate": 0.0004992696605333103, + "loss": 1.9684, + "step": 555 + }, + { + "epoch": 0.054296875, + "grad_norm": 0.28167852759361267, + "learning_rate": 0.0004992639242803492, + "loss": 1.97, + "step": 556 + }, + { + "epoch": 0.05439453125, + "grad_norm": 0.26942357420921326, + "learning_rate": 0.0004992581656254087, + "loss": 1.9825, + "step": 557 + }, + { + "epoch": 0.0544921875, + "grad_norm": 0.25316697359085083, + "learning_rate": 0.0004992523845690644, + "loss": 1.9698, + "step": 558 + }, + { + "epoch": 0.05458984375, + "grad_norm": 0.29587239027023315, + "learning_rate": 0.0004992465811118934, + "loss": 2.0063, + "step": 559 + }, + { + "epoch": 0.0546875, + "grad_norm": 0.3209396004676819, + "learning_rate": 0.0004992407552544757, + "loss": 2.0002, + "step": 560 + }, + { + "epoch": 0.05478515625, + "grad_norm": 0.2541654109954834, + "learning_rate": 0.0004992349069973931, + "loss": 2.0039, + "step": 561 + }, + { + "epoch": 0.0548828125, + "grad_norm": 0.34102505445480347, + "learning_rate": 0.0004992290363412298, + "loss": 1.9181, + "step": 562 + }, + { + "epoch": 0.05498046875, + "grad_norm": 0.36123788356781006, + "learning_rate": 0.0004992231432865723, + "loss": 2.042, + "step": 563 + }, + { + "epoch": 0.055078125, + "grad_norm": 0.27344051003456116, + "learning_rate": 0.0004992172278340093, + "loss": 2.0236, + "step": 564 + }, + { + "epoch": 0.05517578125, + "grad_norm": 0.34781453013420105, + "learning_rate": 0.0004992112899841315, + "loss": 2.0017, + "step": 565 + }, + { + "epoch": 0.0552734375, + "grad_norm": 0.2822319567203522, + "learning_rate": 0.0004992053297375322, + "loss": 1.9607, + "step": 566 + }, + { + "epoch": 0.05537109375, + "grad_norm": 0.27945882081985474, + "learning_rate": 0.0004991993470948066, + "loss": 1.9707, + "step": 567 + }, + { + "epoch": 0.05546875, + "grad_norm": 0.31535595655441284, + "learning_rate": 0.0004991933420565527, + "loss": 1.9708, + "step": 568 + }, + { + "epoch": 0.05556640625, + "grad_norm": 0.2658466398715973, + "learning_rate": 0.00049918731462337, + "loss": 1.9763, + "step": 569 + }, + { + "epoch": 0.0556640625, + "grad_norm": 0.3618657886981964, + "learning_rate": 0.0004991812647958607, + "loss": 2.0154, + "step": 570 + }, + { + "epoch": 0.05576171875, + "grad_norm": 0.34552058577537537, + "learning_rate": 0.000499175192574629, + "loss": 1.978, + "step": 571 + }, + { + "epoch": 0.055859375, + "grad_norm": 0.32903313636779785, + "learning_rate": 0.0004991690979602817, + "loss": 1.9762, + "step": 572 + }, + { + "epoch": 0.05595703125, + "grad_norm": 0.30814382433891296, + "learning_rate": 0.0004991629809534275, + "loss": 1.9944, + "step": 573 + }, + { + "epoch": 0.0560546875, + "grad_norm": 0.21156492829322815, + "learning_rate": 0.0004991568415546775, + "loss": 1.9919, + "step": 574 + }, + { + "epoch": 0.05615234375, + "grad_norm": 0.32283446192741394, + "learning_rate": 0.0004991506797646446, + "loss": 1.9697, + "step": 575 + }, + { + "epoch": 0.05625, + "grad_norm": 0.2915102541446686, + "learning_rate": 0.0004991444955839447, + "loss": 1.9736, + "step": 576 + }, + { + "epoch": 0.05634765625, + "grad_norm": 0.28996148705482483, + "learning_rate": 0.0004991382890131955, + "loss": 2.002, + "step": 577 + }, + { + "epoch": 0.0564453125, + "grad_norm": 0.3859401345252991, + "learning_rate": 0.0004991320600530168, + "loss": 1.9753, + "step": 578 + }, + { + "epoch": 0.05654296875, + "grad_norm": 0.28169432282447815, + "learning_rate": 0.0004991258087040312, + "loss": 1.9798, + "step": 579 + }, + { + "epoch": 0.056640625, + "grad_norm": 0.30669984221458435, + "learning_rate": 0.0004991195349668626, + "loss": 2.0007, + "step": 580 + }, + { + "epoch": 0.05673828125, + "grad_norm": 0.26832297444343567, + "learning_rate": 0.000499113238842138, + "loss": 2.0146, + "step": 581 + }, + { + "epoch": 0.0568359375, + "grad_norm": 0.25695785880088806, + "learning_rate": 0.0004991069203304865, + "loss": 1.9814, + "step": 582 + }, + { + "epoch": 0.05693359375, + "grad_norm": 0.25700533390045166, + "learning_rate": 0.0004991005794325389, + "loss": 1.9623, + "step": 583 + }, + { + "epoch": 0.05703125, + "grad_norm": 0.20786869525909424, + "learning_rate": 0.0004990942161489288, + "loss": 1.9728, + "step": 584 + }, + { + "epoch": 0.05712890625, + "grad_norm": 0.23988646268844604, + "learning_rate": 0.0004990878304802918, + "loss": 1.9749, + "step": 585 + }, + { + "epoch": 0.0572265625, + "grad_norm": 0.2418268769979477, + "learning_rate": 0.0004990814224272658, + "loss": 1.928, + "step": 586 + }, + { + "epoch": 0.05732421875, + "grad_norm": 0.27739882469177246, + "learning_rate": 0.0004990749919904909, + "loss": 1.964, + "step": 587 + }, + { + "epoch": 0.057421875, + "grad_norm": 0.2773842513561249, + "learning_rate": 0.0004990685391706094, + "loss": 1.9589, + "step": 588 + }, + { + "epoch": 0.05751953125, + "grad_norm": 0.24483682215213776, + "learning_rate": 0.0004990620639682659, + "loss": 1.9965, + "step": 589 + }, + { + "epoch": 0.0576171875, + "grad_norm": 0.27345889806747437, + "learning_rate": 0.0004990555663841071, + "loss": 1.9444, + "step": 590 + }, + { + "epoch": 0.05771484375, + "grad_norm": 0.26006460189819336, + "learning_rate": 0.0004990490464187824, + "loss": 1.9391, + "step": 591 + }, + { + "epoch": 0.0578125, + "grad_norm": 0.1935306191444397, + "learning_rate": 0.0004990425040729427, + "loss": 1.9679, + "step": 592 + }, + { + "epoch": 0.05791015625, + "grad_norm": 0.26823100447654724, + "learning_rate": 0.0004990359393472418, + "loss": 2.0055, + "step": 593 + }, + { + "epoch": 0.0580078125, + "grad_norm": 0.2706959545612335, + "learning_rate": 0.0004990293522423352, + "loss": 1.9853, + "step": 594 + }, + { + "epoch": 0.05810546875, + "grad_norm": 0.2099694162607193, + "learning_rate": 0.0004990227427588811, + "loss": 1.959, + "step": 595 + }, + { + "epoch": 0.058203125, + "grad_norm": 0.22345881164073944, + "learning_rate": 0.0004990161108975398, + "loss": 1.9867, + "step": 596 + }, + { + "epoch": 0.05830078125, + "grad_norm": 0.34050315618515015, + "learning_rate": 0.0004990094566589734, + "loss": 1.9555, + "step": 597 + }, + { + "epoch": 0.0583984375, + "grad_norm": 0.5298358798027039, + "learning_rate": 0.0004990027800438468, + "loss": 1.9824, + "step": 598 + }, + { + "epoch": 0.05849609375, + "grad_norm": 0.6408056616783142, + "learning_rate": 0.0004989960810528271, + "loss": 1.9647, + "step": 599 + }, + { + "epoch": 0.05859375, + "grad_norm": 0.4058885872364044, + "learning_rate": 0.0004989893596865833, + "loss": 1.9564, + "step": 600 + }, + { + "epoch": 0.05869140625, + "grad_norm": 0.2911423146724701, + "learning_rate": 0.0004989826159457869, + "loss": 1.9485, + "step": 601 + }, + { + "epoch": 0.0587890625, + "grad_norm": 0.40600210428237915, + "learning_rate": 0.0004989758498311114, + "loss": 1.993, + "step": 602 + }, + { + "epoch": 0.05888671875, + "grad_norm": 0.3920167088508606, + "learning_rate": 0.0004989690613432327, + "loss": 2.0188, + "step": 603 + }, + { + "epoch": 0.058984375, + "grad_norm": 0.2924116551876068, + "learning_rate": 0.0004989622504828291, + "loss": 1.9749, + "step": 604 + }, + { + "epoch": 0.05908203125, + "grad_norm": 0.33497628569602966, + "learning_rate": 0.0004989554172505807, + "loss": 1.9358, + "step": 605 + }, + { + "epoch": 0.0591796875, + "grad_norm": 0.31751033663749695, + "learning_rate": 0.0004989485616471702, + "loss": 1.9886, + "step": 606 + }, + { + "epoch": 0.05927734375, + "grad_norm": 0.2204602211713791, + "learning_rate": 0.0004989416836732825, + "loss": 1.9715, + "step": 607 + }, + { + "epoch": 0.059375, + "grad_norm": 0.2980511784553528, + "learning_rate": 0.0004989347833296044, + "loss": 1.9479, + "step": 608 + }, + { + "epoch": 0.05947265625, + "grad_norm": 0.2874164879322052, + "learning_rate": 0.0004989278606168253, + "loss": 1.9717, + "step": 609 + }, + { + "epoch": 0.0595703125, + "grad_norm": 0.2526591718196869, + "learning_rate": 0.0004989209155356368, + "loss": 2.0009, + "step": 610 + }, + { + "epoch": 0.05966796875, + "grad_norm": 0.2890442907810211, + "learning_rate": 0.0004989139480867327, + "loss": 1.9862, + "step": 611 + }, + { + "epoch": 0.059765625, + "grad_norm": 0.26463401317596436, + "learning_rate": 0.0004989069582708088, + "loss": 2.0143, + "step": 612 + }, + { + "epoch": 0.05986328125, + "grad_norm": 0.19906644523143768, + "learning_rate": 0.0004988999460885634, + "loss": 1.9845, + "step": 613 + }, + { + "epoch": 0.0599609375, + "grad_norm": 0.28752684593200684, + "learning_rate": 0.000498892911540697, + "loss": 1.9421, + "step": 614 + }, + { + "epoch": 0.06005859375, + "grad_norm": 0.2739415466785431, + "learning_rate": 0.0004988858546279123, + "loss": 1.967, + "step": 615 + }, + { + "epoch": 0.06015625, + "grad_norm": 0.27040478587150574, + "learning_rate": 0.000498878775350914, + "loss": 2.0001, + "step": 616 + }, + { + "epoch": 0.06025390625, + "grad_norm": 0.28345200419425964, + "learning_rate": 0.0004988716737104096, + "loss": 1.9637, + "step": 617 + }, + { + "epoch": 0.0603515625, + "grad_norm": 0.2515316903591156, + "learning_rate": 0.0004988645497071082, + "loss": 1.9511, + "step": 618 + }, + { + "epoch": 0.06044921875, + "grad_norm": 0.22784312069416046, + "learning_rate": 0.0004988574033417216, + "loss": 1.9752, + "step": 619 + }, + { + "epoch": 0.060546875, + "grad_norm": 0.23128612339496613, + "learning_rate": 0.0004988502346149636, + "loss": 1.9803, + "step": 620 + }, + { + "epoch": 0.06064453125, + "grad_norm": 0.22022667527198792, + "learning_rate": 0.0004988430435275503, + "loss": 1.9592, + "step": 621 + }, + { + "epoch": 0.0607421875, + "grad_norm": 0.2470719963312149, + "learning_rate": 0.0004988358300802, + "loss": 2.0063, + "step": 622 + }, + { + "epoch": 0.06083984375, + "grad_norm": 0.22507692873477936, + "learning_rate": 0.0004988285942736332, + "loss": 1.9873, + "step": 623 + }, + { + "epoch": 0.0609375, + "grad_norm": 0.2285120040178299, + "learning_rate": 0.0004988213361085729, + "loss": 1.973, + "step": 624 + }, + { + "epoch": 0.06103515625, + "grad_norm": 0.2436297982931137, + "learning_rate": 0.000498814055585744, + "loss": 1.9786, + "step": 625 + }, + { + "epoch": 0.0611328125, + "grad_norm": 0.2736089825630188, + "learning_rate": 0.0004988067527058737, + "loss": 1.9552, + "step": 626 + }, + { + "epoch": 0.06123046875, + "grad_norm": 0.2813490033149719, + "learning_rate": 0.0004987994274696917, + "loss": 2.0173, + "step": 627 + }, + { + "epoch": 0.061328125, + "grad_norm": 0.3024250864982605, + "learning_rate": 0.0004987920798779294, + "loss": 1.9232, + "step": 628 + }, + { + "epoch": 0.06142578125, + "grad_norm": 0.22795617580413818, + "learning_rate": 0.000498784709931321, + "loss": 1.9981, + "step": 629 + }, + { + "epoch": 0.0615234375, + "grad_norm": 0.2585217356681824, + "learning_rate": 0.0004987773176306026, + "loss": 2.0197, + "step": 630 + }, + { + "epoch": 0.06162109375, + "grad_norm": 0.27120688557624817, + "learning_rate": 0.0004987699029765127, + "loss": 1.961, + "step": 631 + }, + { + "epoch": 0.06171875, + "grad_norm": 0.23527094721794128, + "learning_rate": 0.000498762465969792, + "loss": 1.9996, + "step": 632 + }, + { + "epoch": 0.06181640625, + "grad_norm": 0.2519530653953552, + "learning_rate": 0.0004987550066111832, + "loss": 1.9926, + "step": 633 + }, + { + "epoch": 0.0619140625, + "grad_norm": 0.3199564218521118, + "learning_rate": 0.0004987475249014315, + "loss": 2.0066, + "step": 634 + }, + { + "epoch": 0.06201171875, + "grad_norm": 0.3830641806125641, + "learning_rate": 0.0004987400208412843, + "loss": 1.9368, + "step": 635 + }, + { + "epoch": 0.062109375, + "grad_norm": 0.3777483403682709, + "learning_rate": 0.0004987324944314913, + "loss": 1.9792, + "step": 636 + }, + { + "epoch": 0.06220703125, + "grad_norm": 0.420107901096344, + "learning_rate": 0.0004987249456728041, + "loss": 1.9765, + "step": 637 + }, + { + "epoch": 0.0623046875, + "grad_norm": 0.3748738467693329, + "learning_rate": 0.0004987173745659768, + "loss": 1.9826, + "step": 638 + }, + { + "epoch": 0.06240234375, + "grad_norm": 0.3142974078655243, + "learning_rate": 0.0004987097811117658, + "loss": 1.9393, + "step": 639 + }, + { + "epoch": 0.0625, + "grad_norm": 0.24331647157669067, + "learning_rate": 0.0004987021653109296, + "loss": 1.9305, + "step": 640 + }, + { + "epoch": 0.06259765625, + "grad_norm": 0.2792535424232483, + "learning_rate": 0.0004986945271642289, + "loss": 1.9437, + "step": 641 + }, + { + "epoch": 0.0626953125, + "grad_norm": 0.24984711408615112, + "learning_rate": 0.0004986868666724267, + "loss": 1.9838, + "step": 642 + }, + { + "epoch": 0.06279296875, + "grad_norm": 0.2477482557296753, + "learning_rate": 0.0004986791838362881, + "loss": 1.9981, + "step": 643 + }, + { + "epoch": 0.062890625, + "grad_norm": 0.3012976348400116, + "learning_rate": 0.0004986714786565807, + "loss": 1.9963, + "step": 644 + }, + { + "epoch": 0.06298828125, + "grad_norm": 0.2480735331773758, + "learning_rate": 0.0004986637511340742, + "loss": 1.9567, + "step": 645 + }, + { + "epoch": 0.0630859375, + "grad_norm": 0.20119979977607727, + "learning_rate": 0.0004986560012695403, + "loss": 2.0012, + "step": 646 + }, + { + "epoch": 0.06318359375, + "grad_norm": 0.27725833654403687, + "learning_rate": 0.0004986482290637535, + "loss": 1.9751, + "step": 647 + }, + { + "epoch": 0.06328125, + "grad_norm": 0.27348190546035767, + "learning_rate": 0.0004986404345174898, + "loss": 1.976, + "step": 648 + }, + { + "epoch": 0.06337890625, + "grad_norm": 0.29639676213264465, + "learning_rate": 0.000498632617631528, + "loss": 1.9835, + "step": 649 + }, + { + "epoch": 0.0634765625, + "grad_norm": 0.2911996841430664, + "learning_rate": 0.000498624778406649, + "loss": 2.0107, + "step": 650 + }, + { + "epoch": 0.06357421875, + "grad_norm": 0.30200451612472534, + "learning_rate": 0.0004986169168436355, + "loss": 1.9433, + "step": 651 + }, + { + "epoch": 0.063671875, + "grad_norm": 0.288396954536438, + "learning_rate": 0.0004986090329432734, + "loss": 1.9759, + "step": 652 + }, + { + "epoch": 0.06376953125, + "grad_norm": 0.26006969809532166, + "learning_rate": 0.0004986011267063497, + "loss": 1.977, + "step": 653 + }, + { + "epoch": 0.0638671875, + "grad_norm": 0.2467355877161026, + "learning_rate": 0.0004985931981336544, + "loss": 1.9892, + "step": 654 + }, + { + "epoch": 0.06396484375, + "grad_norm": 0.2180376946926117, + "learning_rate": 0.0004985852472259797, + "loss": 2.0012, + "step": 655 + }, + { + "epoch": 0.0640625, + "grad_norm": 0.23811106383800507, + "learning_rate": 0.0004985772739841193, + "loss": 1.969, + "step": 656 + }, + { + "epoch": 0.06416015625, + "grad_norm": 0.2517051696777344, + "learning_rate": 0.00049856927840887, + "loss": 1.9607, + "step": 657 + }, + { + "epoch": 0.0642578125, + "grad_norm": 0.2900746464729309, + "learning_rate": 0.0004985612605010305, + "loss": 1.9865, + "step": 658 + }, + { + "epoch": 0.06435546875, + "grad_norm": 0.2779560983181, + "learning_rate": 0.0004985532202614017, + "loss": 1.9787, + "step": 659 + }, + { + "epoch": 0.064453125, + "grad_norm": 0.2686167061328888, + "learning_rate": 0.0004985451576907865, + "loss": 1.9279, + "step": 660 + }, + { + "epoch": 0.06455078125, + "grad_norm": 0.25136813521385193, + "learning_rate": 0.0004985370727899907, + "loss": 1.9502, + "step": 661 + }, + { + "epoch": 0.0646484375, + "grad_norm": 0.26508840918540955, + "learning_rate": 0.0004985289655598216, + "loss": 1.9588, + "step": 662 + }, + { + "epoch": 0.06474609375, + "grad_norm": 0.2178860753774643, + "learning_rate": 0.000498520836001089, + "loss": 1.9445, + "step": 663 + }, + { + "epoch": 0.06484375, + "grad_norm": 0.21833141148090363, + "learning_rate": 0.0004985126841146052, + "loss": 1.9963, + "step": 664 + }, + { + "epoch": 0.06494140625, + "grad_norm": 0.21572785079479218, + "learning_rate": 0.0004985045099011844, + "loss": 1.9789, + "step": 665 + }, + { + "epoch": 0.0650390625, + "grad_norm": 0.23386414349079132, + "learning_rate": 0.0004984963133616432, + "loss": 1.9829, + "step": 666 + }, + { + "epoch": 0.06513671875, + "grad_norm": 0.25145047903060913, + "learning_rate": 0.0004984880944968003, + "loss": 2.0106, + "step": 667 + }, + { + "epoch": 0.065234375, + "grad_norm": 0.23240795731544495, + "learning_rate": 0.0004984798533074767, + "loss": 1.999, + "step": 668 + }, + { + "epoch": 0.06533203125, + "grad_norm": 0.22223520278930664, + "learning_rate": 0.0004984715897944954, + "loss": 1.9872, + "step": 669 + }, + { + "epoch": 0.0654296875, + "grad_norm": 0.2768714725971222, + "learning_rate": 0.0004984633039586823, + "loss": 1.9656, + "step": 670 + }, + { + "epoch": 0.06552734375, + "grad_norm": 0.35761523246765137, + "learning_rate": 0.0004984549958008646, + "loss": 1.9984, + "step": 671 + }, + { + "epoch": 0.065625, + "grad_norm": 0.4054514169692993, + "learning_rate": 0.0004984466653218726, + "loss": 1.9406, + "step": 672 + }, + { + "epoch": 0.06572265625, + "grad_norm": 0.40501922369003296, + "learning_rate": 0.0004984383125225383, + "loss": 1.9682, + "step": 673 + }, + { + "epoch": 0.0658203125, + "grad_norm": 0.31225842237472534, + "learning_rate": 0.0004984299374036961, + "loss": 1.9833, + "step": 674 + }, + { + "epoch": 0.06591796875, + "grad_norm": 0.24159114062786102, + "learning_rate": 0.0004984215399661825, + "loss": 1.9634, + "step": 675 + }, + { + "epoch": 0.066015625, + "grad_norm": 0.2537291944026947, + "learning_rate": 0.0004984131202108364, + "loss": 1.9563, + "step": 676 + }, + { + "epoch": 0.06611328125, + "grad_norm": 0.29919305443763733, + "learning_rate": 0.0004984046781384988, + "loss": 1.9594, + "step": 677 + }, + { + "epoch": 0.0662109375, + "grad_norm": 0.28458645939826965, + "learning_rate": 0.0004983962137500133, + "loss": 1.9313, + "step": 678 + }, + { + "epoch": 0.06630859375, + "grad_norm": 0.2523960471153259, + "learning_rate": 0.0004983877270462249, + "loss": 1.9616, + "step": 679 + }, + { + "epoch": 0.06640625, + "grad_norm": 0.2605716288089752, + "learning_rate": 0.0004983792180279816, + "loss": 1.9631, + "step": 680 + }, + { + "epoch": 0.06650390625, + "grad_norm": 0.280000239610672, + "learning_rate": 0.0004983706866961335, + "loss": 1.9945, + "step": 681 + }, + { + "epoch": 0.0666015625, + "grad_norm": 0.2881770730018616, + "learning_rate": 0.0004983621330515326, + "loss": 1.9639, + "step": 682 + }, + { + "epoch": 0.06669921875, + "grad_norm": 0.27240094542503357, + "learning_rate": 0.0004983535570950334, + "loss": 1.9226, + "step": 683 + }, + { + "epoch": 0.066796875, + "grad_norm": 0.2580428123474121, + "learning_rate": 0.0004983449588274926, + "loss": 1.9625, + "step": 684 + }, + { + "epoch": 0.06689453125, + "grad_norm": 0.2719865143299103, + "learning_rate": 0.0004983363382497692, + "loss": 1.9941, + "step": 685 + }, + { + "epoch": 0.0669921875, + "grad_norm": 0.2593652904033661, + "learning_rate": 0.000498327695362724, + "loss": 1.9558, + "step": 686 + }, + { + "epoch": 0.06708984375, + "grad_norm": 0.2575826346874237, + "learning_rate": 0.0004983190301672207, + "loss": 2.0188, + "step": 687 + }, + { + "epoch": 0.0671875, + "grad_norm": 0.23726949095726013, + "learning_rate": 0.0004983103426641246, + "loss": 1.9568, + "step": 688 + }, + { + "epoch": 0.06728515625, + "grad_norm": 0.3027707040309906, + "learning_rate": 0.0004983016328543037, + "loss": 1.9708, + "step": 689 + }, + { + "epoch": 0.0673828125, + "grad_norm": 0.29431256651878357, + "learning_rate": 0.0004982929007386279, + "loss": 1.9668, + "step": 690 + }, + { + "epoch": 0.06748046875, + "grad_norm": 0.24073566496372223, + "learning_rate": 0.0004982841463179694, + "loss": 1.9746, + "step": 691 + }, + { + "epoch": 0.067578125, + "grad_norm": 0.27540677785873413, + "learning_rate": 0.000498275369593203, + "loss": 1.868, + "step": 692 + }, + { + "epoch": 0.06767578125, + "grad_norm": 0.2665644884109497, + "learning_rate": 0.0004982665705652052, + "loss": 1.9528, + "step": 693 + }, + { + "epoch": 0.0677734375, + "grad_norm": 0.2117265909910202, + "learning_rate": 0.0004982577492348549, + "loss": 1.9977, + "step": 694 + }, + { + "epoch": 0.06787109375, + "grad_norm": 0.30092543363571167, + "learning_rate": 0.0004982489056030334, + "loss": 1.9636, + "step": 695 + }, + { + "epoch": 0.06796875, + "grad_norm": 0.2956363558769226, + "learning_rate": 0.0004982400396706239, + "loss": 1.9519, + "step": 696 + }, + { + "epoch": 0.06806640625, + "grad_norm": 0.3209246098995209, + "learning_rate": 0.0004982311514385124, + "loss": 1.9761, + "step": 697 + }, + { + "epoch": 0.0681640625, + "grad_norm": 0.4527253806591034, + "learning_rate": 0.0004982222409075863, + "loss": 1.9405, + "step": 698 + }, + { + "epoch": 0.06826171875, + "grad_norm": 0.4838014245033264, + "learning_rate": 0.000498213308078736, + "loss": 1.9434, + "step": 699 + }, + { + "epoch": 0.068359375, + "grad_norm": 0.4865509569644928, + "learning_rate": 0.0004982043529528536, + "loss": 1.9706, + "step": 700 + }, + { + "epoch": 0.06845703125, + "grad_norm": 0.38092029094696045, + "learning_rate": 0.0004981953755308339, + "loss": 1.9168, + "step": 701 + }, + { + "epoch": 0.0685546875, + "grad_norm": 0.24092066287994385, + "learning_rate": 0.0004981863758135735, + "loss": 1.9693, + "step": 702 + }, + { + "epoch": 0.06865234375, + "grad_norm": 0.3893338143825531, + "learning_rate": 0.0004981773538019714, + "loss": 1.9812, + "step": 703 + }, + { + "epoch": 0.06875, + "grad_norm": 0.3019930422306061, + "learning_rate": 0.0004981683094969287, + "loss": 1.9207, + "step": 704 + }, + { + "epoch": 0.06884765625, + "grad_norm": 0.2746503949165344, + "learning_rate": 0.0004981592428993492, + "loss": 1.9416, + "step": 705 + }, + { + "epoch": 0.0689453125, + "grad_norm": 0.2998049557209015, + "learning_rate": 0.0004981501540101382, + "loss": 1.9685, + "step": 706 + }, + { + "epoch": 0.06904296875, + "grad_norm": 0.2986246645450592, + "learning_rate": 0.0004981410428302037, + "loss": 1.9827, + "step": 707 + }, + { + "epoch": 0.069140625, + "grad_norm": 0.28714433312416077, + "learning_rate": 0.000498131909360456, + "loss": 1.904, + "step": 708 + }, + { + "epoch": 0.06923828125, + "grad_norm": 0.21930216252803802, + "learning_rate": 0.0004981227536018073, + "loss": 1.966, + "step": 709 + }, + { + "epoch": 0.0693359375, + "grad_norm": 0.2309812754392624, + "learning_rate": 0.0004981135755551722, + "loss": 1.9983, + "step": 710 + }, + { + "epoch": 0.06943359375, + "grad_norm": 0.2016872614622116, + "learning_rate": 0.0004981043752214674, + "loss": 1.9382, + "step": 711 + }, + { + "epoch": 0.06953125, + "grad_norm": 0.235287606716156, + "learning_rate": 0.0004980951526016122, + "loss": 1.9914, + "step": 712 + }, + { + "epoch": 0.06962890625, + "grad_norm": 0.23780131340026855, + "learning_rate": 0.0004980859076965276, + "loss": 1.9404, + "step": 713 + }, + { + "epoch": 0.0697265625, + "grad_norm": 0.2195412665605545, + "learning_rate": 0.0004980766405071372, + "loss": 1.9526, + "step": 714 + }, + { + "epoch": 0.06982421875, + "grad_norm": 0.2391882836818695, + "learning_rate": 0.0004980673510343668, + "loss": 1.9179, + "step": 715 + }, + { + "epoch": 0.069921875, + "grad_norm": 0.27623385190963745, + "learning_rate": 0.000498058039279144, + "loss": 1.9519, + "step": 716 + }, + { + "epoch": 0.07001953125, + "grad_norm": 0.27679604291915894, + "learning_rate": 0.0004980487052423994, + "loss": 2.0292, + "step": 717 + }, + { + "epoch": 0.0701171875, + "grad_norm": 0.2968669533729553, + "learning_rate": 0.0004980393489250651, + "loss": 1.9162, + "step": 718 + }, + { + "epoch": 0.07021484375, + "grad_norm": 0.2479199320077896, + "learning_rate": 0.0004980299703280758, + "loss": 1.9544, + "step": 719 + }, + { + "epoch": 0.0703125, + "grad_norm": 0.2597293555736542, + "learning_rate": 0.0004980205694523682, + "loss": 1.9273, + "step": 720 + }, + { + "epoch": 0.07041015625, + "grad_norm": 0.23868051171302795, + "learning_rate": 0.0004980111462988817, + "loss": 1.9527, + "step": 721 + }, + { + "epoch": 0.0705078125, + "grad_norm": 0.28150075674057007, + "learning_rate": 0.0004980017008685573, + "loss": 1.9523, + "step": 722 + }, + { + "epoch": 0.07060546875, + "grad_norm": 0.3689873516559601, + "learning_rate": 0.0004979922331623387, + "loss": 1.9167, + "step": 723 + }, + { + "epoch": 0.070703125, + "grad_norm": 0.380206823348999, + "learning_rate": 0.0004979827431811713, + "loss": 1.9025, + "step": 724 + }, + { + "epoch": 0.07080078125, + "grad_norm": 0.2919938266277313, + "learning_rate": 0.0004979732309260036, + "loss": 1.9996, + "step": 725 + }, + { + "epoch": 0.0708984375, + "grad_norm": 0.302680104970932, + "learning_rate": 0.0004979636963977853, + "loss": 1.95, + "step": 726 + }, + { + "epoch": 0.07099609375, + "grad_norm": 0.25307831168174744, + "learning_rate": 0.0004979541395974691, + "loss": 1.9723, + "step": 727 + }, + { + "epoch": 0.07109375, + "grad_norm": 0.304147869348526, + "learning_rate": 0.0004979445605260096, + "loss": 2.0359, + "step": 728 + }, + { + "epoch": 0.07119140625, + "grad_norm": 0.35432276129722595, + "learning_rate": 0.0004979349591843635, + "loss": 1.9384, + "step": 729 + }, + { + "epoch": 0.0712890625, + "grad_norm": 0.23866906762123108, + "learning_rate": 0.00049792533557349, + "loss": 1.9341, + "step": 730 + }, + { + "epoch": 0.07138671875, + "grad_norm": 0.30183711647987366, + "learning_rate": 0.0004979156896943504, + "loss": 1.9603, + "step": 731 + }, + { + "epoch": 0.071484375, + "grad_norm": 0.30881768465042114, + "learning_rate": 0.0004979060215479083, + "loss": 1.982, + "step": 732 + }, + { + "epoch": 0.07158203125, + "grad_norm": 0.2996484041213989, + "learning_rate": 0.0004978963311351295, + "loss": 1.9536, + "step": 733 + }, + { + "epoch": 0.0716796875, + "grad_norm": 0.33594587445259094, + "learning_rate": 0.0004978866184569818, + "loss": 1.9813, + "step": 734 + }, + { + "epoch": 0.07177734375, + "grad_norm": 0.30377864837646484, + "learning_rate": 0.0004978768835144355, + "loss": 1.9417, + "step": 735 + }, + { + "epoch": 0.071875, + "grad_norm": 0.34858009219169617, + "learning_rate": 0.0004978671263084631, + "loss": 1.9368, + "step": 736 + }, + { + "epoch": 0.07197265625, + "grad_norm": 0.28119683265686035, + "learning_rate": 0.0004978573468400392, + "loss": 1.9223, + "step": 737 + }, + { + "epoch": 0.0720703125, + "grad_norm": 0.2478085160255432, + "learning_rate": 0.0004978475451101408, + "loss": 1.9744, + "step": 738 + }, + { + "epoch": 0.07216796875, + "grad_norm": 0.2461906224489212, + "learning_rate": 0.0004978377211197466, + "loss": 1.961, + "step": 739 + }, + { + "epoch": 0.072265625, + "grad_norm": 0.2495601922273636, + "learning_rate": 0.0004978278748698385, + "loss": 1.9602, + "step": 740 + }, + { + "epoch": 0.07236328125, + "grad_norm": 0.3028644323348999, + "learning_rate": 0.0004978180063613997, + "loss": 1.95, + "step": 741 + }, + { + "epoch": 0.0724609375, + "grad_norm": 0.3690338730812073, + "learning_rate": 0.000497808115595416, + "loss": 1.9625, + "step": 742 + }, + { + "epoch": 0.07255859375, + "grad_norm": 0.3791324198246002, + "learning_rate": 0.0004977982025728755, + "loss": 1.9747, + "step": 743 + }, + { + "epoch": 0.07265625, + "grad_norm": 0.29515188932418823, + "learning_rate": 0.0004977882672947683, + "loss": 1.9774, + "step": 744 + }, + { + "epoch": 0.07275390625, + "grad_norm": 0.29207196831703186, + "learning_rate": 0.000497778309762087, + "loss": 1.9704, + "step": 745 + }, + { + "epoch": 0.0728515625, + "grad_norm": 0.23698067665100098, + "learning_rate": 0.0004977683299758262, + "loss": 1.978, + "step": 746 + }, + { + "epoch": 0.07294921875, + "grad_norm": 0.2712165415287018, + "learning_rate": 0.0004977583279369827, + "loss": 1.9524, + "step": 747 + }, + { + "epoch": 0.073046875, + "grad_norm": 0.32161638140678406, + "learning_rate": 0.0004977483036465558, + "loss": 1.9532, + "step": 748 + }, + { + "epoch": 0.07314453125, + "grad_norm": 0.25917574763298035, + "learning_rate": 0.0004977382571055467, + "loss": 1.9307, + "step": 749 + }, + { + "epoch": 0.0732421875, + "grad_norm": 0.27562615275382996, + "learning_rate": 0.000497728188314959, + "loss": 1.9842, + "step": 750 + }, + { + "epoch": 0.07333984375, + "grad_norm": 0.23242716491222382, + "learning_rate": 0.0004977180972757986, + "loss": 1.9097, + "step": 751 + }, + { + "epoch": 0.0734375, + "grad_norm": 0.3111313283443451, + "learning_rate": 0.0004977079839890732, + "loss": 1.9812, + "step": 752 + }, + { + "epoch": 0.07353515625, + "grad_norm": 0.3295767605304718, + "learning_rate": 0.0004976978484557933, + "loss": 1.9206, + "step": 753 + }, + { + "epoch": 0.0736328125, + "grad_norm": 0.3592395782470703, + "learning_rate": 0.0004976876906769713, + "loss": 1.9817, + "step": 754 + }, + { + "epoch": 0.07373046875, + "grad_norm": 0.3530953824520111, + "learning_rate": 0.0004976775106536218, + "loss": 1.9831, + "step": 755 + }, + { + "epoch": 0.073828125, + "grad_norm": 0.3064524829387665, + "learning_rate": 0.0004976673083867617, + "loss": 1.9207, + "step": 756 + }, + { + "epoch": 0.07392578125, + "grad_norm": 0.281597763299942, + "learning_rate": 0.0004976570838774102, + "loss": 1.9781, + "step": 757 + }, + { + "epoch": 0.0740234375, + "grad_norm": 0.296041876077652, + "learning_rate": 0.0004976468371265886, + "loss": 1.9708, + "step": 758 + }, + { + "epoch": 0.07412109375, + "grad_norm": 0.3518696427345276, + "learning_rate": 0.0004976365681353203, + "loss": 1.9333, + "step": 759 + }, + { + "epoch": 0.07421875, + "grad_norm": 0.21943148970603943, + "learning_rate": 0.0004976262769046313, + "loss": 1.9416, + "step": 760 + }, + { + "epoch": 0.07431640625, + "grad_norm": 0.2540123760700226, + "learning_rate": 0.0004976159634355495, + "loss": 1.989, + "step": 761 + }, + { + "epoch": 0.0744140625, + "grad_norm": 0.3120715320110321, + "learning_rate": 0.0004976056277291053, + "loss": 1.95, + "step": 762 + }, + { + "epoch": 0.07451171875, + "grad_norm": 0.23797550797462463, + "learning_rate": 0.0004975952697863308, + "loss": 1.9534, + "step": 763 + }, + { + "epoch": 0.074609375, + "grad_norm": 0.30815449357032776, + "learning_rate": 0.000497584889608261, + "loss": 2.0283, + "step": 764 + }, + { + "epoch": 0.07470703125, + "grad_norm": 0.37625932693481445, + "learning_rate": 0.0004975744871959327, + "loss": 1.9658, + "step": 765 + }, + { + "epoch": 0.0748046875, + "grad_norm": 0.27407678961753845, + "learning_rate": 0.0004975640625503848, + "loss": 1.9458, + "step": 766 + }, + { + "epoch": 0.07490234375, + "grad_norm": 0.26577049493789673, + "learning_rate": 0.000497553615672659, + "loss": 1.9667, + "step": 767 + }, + { + "epoch": 0.075, + "grad_norm": 0.29821062088012695, + "learning_rate": 0.0004975431465637983, + "loss": 1.9805, + "step": 768 + }, + { + "epoch": 0.07509765625, + "grad_norm": 0.2582527995109558, + "learning_rate": 0.0004975326552248491, + "loss": 1.9904, + "step": 769 + }, + { + "epoch": 0.0751953125, + "grad_norm": 0.2533649802207947, + "learning_rate": 0.0004975221416568591, + "loss": 1.9383, + "step": 770 + }, + { + "epoch": 0.07529296875, + "grad_norm": 0.253592312335968, + "learning_rate": 0.0004975116058608784, + "loss": 1.9855, + "step": 771 + }, + { + "epoch": 0.075390625, + "grad_norm": 0.2658921182155609, + "learning_rate": 0.0004975010478379596, + "loss": 1.9845, + "step": 772 + }, + { + "epoch": 0.07548828125, + "grad_norm": 0.25861868262290955, + "learning_rate": 0.0004974904675891573, + "loss": 1.9776, + "step": 773 + }, + { + "epoch": 0.0755859375, + "grad_norm": 0.272682249546051, + "learning_rate": 0.0004974798651155284, + "loss": 1.9357, + "step": 774 + }, + { + "epoch": 0.07568359375, + "grad_norm": 0.2574002742767334, + "learning_rate": 0.0004974692404181319, + "loss": 1.9677, + "step": 775 + }, + { + "epoch": 0.07578125, + "grad_norm": 0.3008319139480591, + "learning_rate": 0.0004974585934980293, + "loss": 1.983, + "step": 776 + }, + { + "epoch": 0.07587890625, + "grad_norm": 0.3242790699005127, + "learning_rate": 0.0004974479243562839, + "loss": 1.9846, + "step": 777 + }, + { + "epoch": 0.0759765625, + "grad_norm": 0.29044070839881897, + "learning_rate": 0.0004974372329939617, + "loss": 2.0012, + "step": 778 + }, + { + "epoch": 0.07607421875, + "grad_norm": 0.32275035977363586, + "learning_rate": 0.0004974265194121303, + "loss": 1.9842, + "step": 779 + }, + { + "epoch": 0.076171875, + "grad_norm": 0.31752893328666687, + "learning_rate": 0.0004974157836118604, + "loss": 1.9368, + "step": 780 + }, + { + "epoch": 0.07626953125, + "grad_norm": 0.27947962284088135, + "learning_rate": 0.0004974050255942239, + "loss": 1.9794, + "step": 781 + }, + { + "epoch": 0.0763671875, + "grad_norm": 0.29937905073165894, + "learning_rate": 0.000497394245360296, + "loss": 1.9597, + "step": 782 + }, + { + "epoch": 0.07646484375, + "grad_norm": 0.2969702482223511, + "learning_rate": 0.0004973834429111529, + "loss": 1.945, + "step": 783 + }, + { + "epoch": 0.0765625, + "grad_norm": 0.3729354441165924, + "learning_rate": 0.0004973726182478741, + "loss": 1.9032, + "step": 784 + }, + { + "epoch": 0.07666015625, + "grad_norm": 0.32856106758117676, + "learning_rate": 0.0004973617713715408, + "loss": 1.9551, + "step": 785 + }, + { + "epoch": 0.0767578125, + "grad_norm": 0.23239392042160034, + "learning_rate": 0.0004973509022832364, + "loss": 1.9632, + "step": 786 + }, + { + "epoch": 0.07685546875, + "grad_norm": 0.23353087902069092, + "learning_rate": 0.000497340010984047, + "loss": 1.9514, + "step": 787 + }, + { + "epoch": 0.076953125, + "grad_norm": 0.23280592262744904, + "learning_rate": 0.00049732909747506, + "loss": 1.9696, + "step": 788 + }, + { + "epoch": 0.07705078125, + "grad_norm": 0.2193138748407364, + "learning_rate": 0.0004973181617573658, + "loss": 1.9734, + "step": 789 + }, + { + "epoch": 0.0771484375, + "grad_norm": 0.3234734833240509, + "learning_rate": 0.0004973072038320569, + "loss": 1.9874, + "step": 790 + }, + { + "epoch": 0.07724609375, + "grad_norm": 0.35144004225730896, + "learning_rate": 0.0004972962237002278, + "loss": 1.9331, + "step": 791 + }, + { + "epoch": 0.07734375, + "grad_norm": 0.37716230750083923, + "learning_rate": 0.0004972852213629753, + "loss": 1.9731, + "step": 792 + }, + { + "epoch": 0.07744140625, + "grad_norm": 0.3221200108528137, + "learning_rate": 0.0004972741968213985, + "loss": 2.0043, + "step": 793 + }, + { + "epoch": 0.0775390625, + "grad_norm": 0.23167745769023895, + "learning_rate": 0.0004972631500765985, + "loss": 1.9634, + "step": 794 + }, + { + "epoch": 0.07763671875, + "grad_norm": 0.4177049398422241, + "learning_rate": 0.0004972520811296789, + "loss": 1.9533, + "step": 795 + }, + { + "epoch": 0.077734375, + "grad_norm": 0.39038336277008057, + "learning_rate": 0.0004972409899817455, + "loss": 1.9508, + "step": 796 + }, + { + "epoch": 0.07783203125, + "grad_norm": 0.4033074676990509, + "learning_rate": 0.0004972298766339061, + "loss": 1.965, + "step": 797 + }, + { + "epoch": 0.0779296875, + "grad_norm": 0.3271097242832184, + "learning_rate": 0.0004972187410872707, + "loss": 1.9893, + "step": 798 + }, + { + "epoch": 0.07802734375, + "grad_norm": 0.30553460121154785, + "learning_rate": 0.0004972075833429516, + "loss": 1.9735, + "step": 799 + }, + { + "epoch": 0.078125, + "grad_norm": 0.28286099433898926, + "learning_rate": 0.0004971964034020637, + "loss": 1.9704, + "step": 800 + }, + { + "epoch": 0.07822265625, + "grad_norm": 0.28064650297164917, + "learning_rate": 0.0004971852012657235, + "loss": 1.9452, + "step": 801 + }, + { + "epoch": 0.0783203125, + "grad_norm": 0.2370154857635498, + "learning_rate": 0.0004971739769350502, + "loss": 2.0068, + "step": 802 + }, + { + "epoch": 0.07841796875, + "grad_norm": 0.26157453656196594, + "learning_rate": 0.0004971627304111648, + "loss": 1.9209, + "step": 803 + }, + { + "epoch": 0.078515625, + "grad_norm": 0.2244575023651123, + "learning_rate": 0.0004971514616951908, + "loss": 1.9952, + "step": 804 + }, + { + "epoch": 0.07861328125, + "grad_norm": 0.241064190864563, + "learning_rate": 0.0004971401707882538, + "loss": 1.9616, + "step": 805 + }, + { + "epoch": 0.0787109375, + "grad_norm": 0.23037207126617432, + "learning_rate": 0.0004971288576914819, + "loss": 1.9792, + "step": 806 + }, + { + "epoch": 0.07880859375, + "grad_norm": 0.2162441909313202, + "learning_rate": 0.0004971175224060049, + "loss": 1.9556, + "step": 807 + }, + { + "epoch": 0.07890625, + "grad_norm": 0.2614160180091858, + "learning_rate": 0.0004971061649329551, + "loss": 1.9247, + "step": 808 + }, + { + "epoch": 0.07900390625, + "grad_norm": 0.22268271446228027, + "learning_rate": 0.0004970947852734672, + "loss": 1.9596, + "step": 809 + }, + { + "epoch": 0.0791015625, + "grad_norm": 0.26265689730644226, + "learning_rate": 0.0004970833834286779, + "loss": 1.9722, + "step": 810 + }, + { + "epoch": 0.07919921875, + "grad_norm": 0.29413190484046936, + "learning_rate": 0.000497071959399726, + "loss": 1.9808, + "step": 811 + }, + { + "epoch": 0.079296875, + "grad_norm": 0.3389902114868164, + "learning_rate": 0.0004970605131877527, + "loss": 1.9986, + "step": 812 + }, + { + "epoch": 0.07939453125, + "grad_norm": 0.3456060290336609, + "learning_rate": 0.0004970490447939016, + "loss": 1.9556, + "step": 813 + }, + { + "epoch": 0.0794921875, + "grad_norm": 0.3039393126964569, + "learning_rate": 0.0004970375542193179, + "loss": 1.9359, + "step": 814 + }, + { + "epoch": 0.07958984375, + "grad_norm": 0.46128201484680176, + "learning_rate": 0.0004970260414651497, + "loss": 1.935, + "step": 815 + }, + { + "epoch": 0.0796875, + "grad_norm": 0.5321182608604431, + "learning_rate": 0.000497014506532547, + "loss": 1.9767, + "step": 816 + }, + { + "epoch": 0.07978515625, + "grad_norm": 0.39562714099884033, + "learning_rate": 0.0004970029494226618, + "loss": 1.9803, + "step": 817 + }, + { + "epoch": 0.0798828125, + "grad_norm": 0.2717466950416565, + "learning_rate": 0.0004969913701366488, + "loss": 1.9675, + "step": 818 + }, + { + "epoch": 0.07998046875, + "grad_norm": 0.32809481024742126, + "learning_rate": 0.0004969797686756646, + "loss": 1.9869, + "step": 819 + }, + { + "epoch": 0.080078125, + "grad_norm": 0.3525324761867523, + "learning_rate": 0.0004969681450408681, + "loss": 1.9858, + "step": 820 + }, + { + "epoch": 0.08017578125, + "grad_norm": 0.30448198318481445, + "learning_rate": 0.0004969564992334202, + "loss": 2.0067, + "step": 821 + }, + { + "epoch": 0.0802734375, + "grad_norm": 0.27088144421577454, + "learning_rate": 0.0004969448312544845, + "loss": 2.0068, + "step": 822 + }, + { + "epoch": 0.08037109375, + "grad_norm": 0.26881691813468933, + "learning_rate": 0.0004969331411052265, + "loss": 1.9418, + "step": 823 + }, + { + "epoch": 0.08046875, + "grad_norm": 0.31323063373565674, + "learning_rate": 0.0004969214287868138, + "loss": 1.9662, + "step": 824 + }, + { + "epoch": 0.08056640625, + "grad_norm": 0.25669077038764954, + "learning_rate": 0.0004969096943004163, + "loss": 1.9733, + "step": 825 + }, + { + "epoch": 0.0806640625, + "grad_norm": 0.27999046444892883, + "learning_rate": 0.0004968979376472065, + "loss": 1.9611, + "step": 826 + }, + { + "epoch": 0.08076171875, + "grad_norm": 0.35361799597740173, + "learning_rate": 0.0004968861588283584, + "loss": 1.9035, + "step": 827 + }, + { + "epoch": 0.080859375, + "grad_norm": 0.290341317653656, + "learning_rate": 0.0004968743578450489, + "loss": 2.0007, + "step": 828 + }, + { + "epoch": 0.08095703125, + "grad_norm": 0.29945844411849976, + "learning_rate": 0.0004968625346984565, + "loss": 1.9724, + "step": 829 + }, + { + "epoch": 0.0810546875, + "grad_norm": 0.2630455493927002, + "learning_rate": 0.0004968506893897626, + "loss": 1.9765, + "step": 830 + }, + { + "epoch": 0.08115234375, + "grad_norm": 0.2183736264705658, + "learning_rate": 0.0004968388219201501, + "loss": 1.9414, + "step": 831 + }, + { + "epoch": 0.08125, + "grad_norm": 0.24703919887542725, + "learning_rate": 0.0004968269322908046, + "loss": 1.9725, + "step": 832 + }, + { + "epoch": 0.08134765625, + "grad_norm": 0.22808632254600525, + "learning_rate": 0.0004968150205029139, + "loss": 1.9121, + "step": 833 + }, + { + "epoch": 0.0814453125, + "grad_norm": 0.18958304822444916, + "learning_rate": 0.0004968030865576677, + "loss": 1.971, + "step": 834 + }, + { + "epoch": 0.08154296875, + "grad_norm": 0.22013933956623077, + "learning_rate": 0.0004967911304562581, + "loss": 1.9703, + "step": 835 + }, + { + "epoch": 0.081640625, + "grad_norm": 0.21636781096458435, + "learning_rate": 0.0004967791521998795, + "loss": 1.992, + "step": 836 + }, + { + "epoch": 0.08173828125, + "grad_norm": 0.20808343589305878, + "learning_rate": 0.0004967671517897284, + "loss": 1.9881, + "step": 837 + }, + { + "epoch": 0.0818359375, + "grad_norm": 0.23455387353897095, + "learning_rate": 0.0004967551292270034, + "loss": 1.9635, + "step": 838 + }, + { + "epoch": 0.08193359375, + "grad_norm": 0.2605065703392029, + "learning_rate": 0.0004967430845129057, + "loss": 1.8877, + "step": 839 + }, + { + "epoch": 0.08203125, + "grad_norm": 0.30832260847091675, + "learning_rate": 0.0004967310176486382, + "loss": 1.9557, + "step": 840 + }, + { + "epoch": 0.08212890625, + "grad_norm": 0.32088643312454224, + "learning_rate": 0.0004967189286354066, + "loss": 1.9352, + "step": 841 + }, + { + "epoch": 0.0822265625, + "grad_norm": 0.3073631823062897, + "learning_rate": 0.0004967068174744181, + "loss": 1.9487, + "step": 842 + }, + { + "epoch": 0.08232421875, + "grad_norm": 0.23467223346233368, + "learning_rate": 0.0004966946841668828, + "loss": 1.9514, + "step": 843 + }, + { + "epoch": 0.082421875, + "grad_norm": 0.2447347342967987, + "learning_rate": 0.0004966825287140125, + "loss": 1.9435, + "step": 844 + }, + { + "epoch": 0.08251953125, + "grad_norm": 0.2715449631214142, + "learning_rate": 0.0004966703511170215, + "loss": 1.9591, + "step": 845 + }, + { + "epoch": 0.0826171875, + "grad_norm": 0.28944098949432373, + "learning_rate": 0.0004966581513771263, + "loss": 1.952, + "step": 846 + }, + { + "epoch": 0.08271484375, + "grad_norm": 0.25582170486450195, + "learning_rate": 0.0004966459294955455, + "loss": 1.9486, + "step": 847 + }, + { + "epoch": 0.0828125, + "grad_norm": 0.305513858795166, + "learning_rate": 0.0004966336854734999, + "loss": 1.9416, + "step": 848 + }, + { + "epoch": 0.08291015625, + "grad_norm": 0.28799137473106384, + "learning_rate": 0.0004966214193122127, + "loss": 1.9417, + "step": 849 + }, + { + "epoch": 0.0830078125, + "grad_norm": 0.22209681570529938, + "learning_rate": 0.000496609131012909, + "loss": 1.9641, + "step": 850 + }, + { + "epoch": 0.08310546875, + "grad_norm": 0.35411134362220764, + "learning_rate": 0.0004965968205768165, + "loss": 1.9497, + "step": 851 + }, + { + "epoch": 0.083203125, + "grad_norm": 0.2821413576602936, + "learning_rate": 0.0004965844880051647, + "loss": 1.9398, + "step": 852 + }, + { + "epoch": 0.08330078125, + "grad_norm": 0.31592175364494324, + "learning_rate": 0.0004965721332991856, + "loss": 1.9407, + "step": 853 + }, + { + "epoch": 0.0833984375, + "grad_norm": 0.3579123318195343, + "learning_rate": 0.0004965597564601134, + "loss": 1.9767, + "step": 854 + }, + { + "epoch": 0.08349609375, + "grad_norm": 0.3371833860874176, + "learning_rate": 0.0004965473574891845, + "loss": 2.0025, + "step": 855 + }, + { + "epoch": 0.08359375, + "grad_norm": 0.39282771944999695, + "learning_rate": 0.0004965349363876371, + "loss": 1.9303, + "step": 856 + }, + { + "epoch": 0.08369140625, + "grad_norm": 0.34869179129600525, + "learning_rate": 0.0004965224931567122, + "loss": 1.9613, + "step": 857 + }, + { + "epoch": 0.0837890625, + "grad_norm": 0.2918208837509155, + "learning_rate": 0.0004965100277976528, + "loss": 1.9242, + "step": 858 + }, + { + "epoch": 0.08388671875, + "grad_norm": 0.25440847873687744, + "learning_rate": 0.0004964975403117041, + "loss": 1.9223, + "step": 859 + }, + { + "epoch": 0.083984375, + "grad_norm": 0.27069318294525146, + "learning_rate": 0.0004964850307001132, + "loss": 1.9711, + "step": 860 + }, + { + "epoch": 0.08408203125, + "grad_norm": 0.27405187487602234, + "learning_rate": 0.0004964724989641301, + "loss": 1.9646, + "step": 861 + }, + { + "epoch": 0.0841796875, + "grad_norm": 0.3116970956325531, + "learning_rate": 0.0004964599451050064, + "loss": 1.9231, + "step": 862 + }, + { + "epoch": 0.08427734375, + "grad_norm": 0.3815697729587555, + "learning_rate": 0.000496447369123996, + "loss": 1.9773, + "step": 863 + }, + { + "epoch": 0.084375, + "grad_norm": 0.32103797793388367, + "learning_rate": 0.0004964347710223555, + "loss": 1.936, + "step": 864 + }, + { + "epoch": 0.08447265625, + "grad_norm": 0.42826011776924133, + "learning_rate": 0.0004964221508013429, + "loss": 1.9315, + "step": 865 + }, + { + "epoch": 0.0845703125, + "grad_norm": 0.3091580271720886, + "learning_rate": 0.0004964095084622192, + "loss": 1.9362, + "step": 866 + }, + { + "epoch": 0.08466796875, + "grad_norm": 0.2816343605518341, + "learning_rate": 0.0004963968440062471, + "loss": 2.0076, + "step": 867 + }, + { + "epoch": 0.084765625, + "grad_norm": 0.35303476452827454, + "learning_rate": 0.0004963841574346917, + "loss": 1.9399, + "step": 868 + }, + { + "epoch": 0.08486328125, + "grad_norm": 0.27021846175193787, + "learning_rate": 0.0004963714487488202, + "loss": 1.9556, + "step": 869 + }, + { + "epoch": 0.0849609375, + "grad_norm": 0.28174805641174316, + "learning_rate": 0.0004963587179499021, + "loss": 1.9749, + "step": 870 + }, + { + "epoch": 0.08505859375, + "grad_norm": 0.2774275243282318, + "learning_rate": 0.0004963459650392093, + "loss": 1.9457, + "step": 871 + }, + { + "epoch": 0.08515625, + "grad_norm": 0.2178190052509308, + "learning_rate": 0.0004963331900180154, + "loss": 2.0192, + "step": 872 + }, + { + "epoch": 0.08525390625, + "grad_norm": 0.2983119487762451, + "learning_rate": 0.0004963203928875967, + "loss": 1.9659, + "step": 873 + }, + { + "epoch": 0.0853515625, + "grad_norm": 0.2878938913345337, + "learning_rate": 0.0004963075736492315, + "loss": 1.9581, + "step": 874 + }, + { + "epoch": 0.08544921875, + "grad_norm": 0.19974899291992188, + "learning_rate": 0.0004962947323042002, + "loss": 1.9898, + "step": 875 + }, + { + "epoch": 0.085546875, + "grad_norm": 0.31860581040382385, + "learning_rate": 0.0004962818688537857, + "loss": 1.9737, + "step": 876 + }, + { + "epoch": 0.08564453125, + "grad_norm": 0.2651256322860718, + "learning_rate": 0.0004962689832992729, + "loss": 1.9977, + "step": 877 + }, + { + "epoch": 0.0857421875, + "grad_norm": 0.286821573972702, + "learning_rate": 0.0004962560756419489, + "loss": 1.9317, + "step": 878 + }, + { + "epoch": 0.08583984375, + "grad_norm": 0.3837340772151947, + "learning_rate": 0.0004962431458831032, + "loss": 1.9419, + "step": 879 + }, + { + "epoch": 0.0859375, + "grad_norm": 0.32222527265548706, + "learning_rate": 0.0004962301940240272, + "loss": 1.9603, + "step": 880 + }, + { + "epoch": 0.08603515625, + "grad_norm": 0.33181512355804443, + "learning_rate": 0.0004962172200660146, + "loss": 1.9646, + "step": 881 + }, + { + "epoch": 0.0861328125, + "grad_norm": 0.2813611626625061, + "learning_rate": 0.0004962042240103617, + "loss": 1.9706, + "step": 882 + }, + { + "epoch": 0.08623046875, + "grad_norm": 0.2933270335197449, + "learning_rate": 0.0004961912058583664, + "loss": 1.9529, + "step": 883 + }, + { + "epoch": 0.086328125, + "grad_norm": 0.31641921401023865, + "learning_rate": 0.0004961781656113292, + "loss": 1.9437, + "step": 884 + }, + { + "epoch": 0.08642578125, + "grad_norm": 0.39271247386932373, + "learning_rate": 0.0004961651032705528, + "loss": 1.957, + "step": 885 + }, + { + "epoch": 0.0865234375, + "grad_norm": 0.3720405697822571, + "learning_rate": 0.000496152018837342, + "loss": 1.9393, + "step": 886 + }, + { + "epoch": 0.08662109375, + "grad_norm": 0.27104172110557556, + "learning_rate": 0.0004961389123130037, + "loss": 1.9547, + "step": 887 + }, + { + "epoch": 0.08671875, + "grad_norm": 0.24182896316051483, + "learning_rate": 0.0004961257836988472, + "loss": 1.9456, + "step": 888 + }, + { + "epoch": 0.08681640625, + "grad_norm": 0.22501450777053833, + "learning_rate": 0.0004961126329961838, + "loss": 1.9336, + "step": 889 + }, + { + "epoch": 0.0869140625, + "grad_norm": 0.2969515323638916, + "learning_rate": 0.0004960994602063273, + "loss": 1.9415, + "step": 890 + }, + { + "epoch": 0.08701171875, + "grad_norm": 0.3117569386959076, + "learning_rate": 0.0004960862653305934, + "loss": 1.8993, + "step": 891 + }, + { + "epoch": 0.087109375, + "grad_norm": 0.32299551367759705, + "learning_rate": 0.0004960730483703004, + "loss": 1.9879, + "step": 892 + }, + { + "epoch": 0.08720703125, + "grad_norm": 0.335592657327652, + "learning_rate": 0.0004960598093267682, + "loss": 1.9665, + "step": 893 + }, + { + "epoch": 0.0873046875, + "grad_norm": 0.21160855889320374, + "learning_rate": 0.0004960465482013197, + "loss": 1.8927, + "step": 894 + }, + { + "epoch": 0.08740234375, + "grad_norm": 0.30621418356895447, + "learning_rate": 0.0004960332649952792, + "loss": 1.9828, + "step": 895 + }, + { + "epoch": 0.0875, + "grad_norm": 0.3541114330291748, + "learning_rate": 0.0004960199597099737, + "loss": 1.9478, + "step": 896 + }, + { + "epoch": 0.08759765625, + "grad_norm": 0.31596067547798157, + "learning_rate": 0.0004960066323467323, + "loss": 1.9355, + "step": 897 + }, + { + "epoch": 0.0876953125, + "grad_norm": 0.3170725107192993, + "learning_rate": 0.0004959932829068863, + "loss": 1.9501, + "step": 898 + }, + { + "epoch": 0.08779296875, + "grad_norm": 0.26953962445259094, + "learning_rate": 0.0004959799113917691, + "loss": 1.9488, + "step": 899 + }, + { + "epoch": 0.087890625, + "grad_norm": 0.23792020976543427, + "learning_rate": 0.0004959665178027165, + "loss": 1.941, + "step": 900 + }, + { + "epoch": 0.08798828125, + "grad_norm": 0.2375611513853073, + "learning_rate": 0.0004959531021410664, + "loss": 1.9133, + "step": 901 + }, + { + "epoch": 0.0880859375, + "grad_norm": 0.20919479429721832, + "learning_rate": 0.0004959396644081588, + "loss": 1.9624, + "step": 902 + }, + { + "epoch": 0.08818359375, + "grad_norm": 0.22488316893577576, + "learning_rate": 0.0004959262046053361, + "loss": 1.9408, + "step": 903 + }, + { + "epoch": 0.08828125, + "grad_norm": 0.2249651849269867, + "learning_rate": 0.0004959127227339427, + "loss": 2.0022, + "step": 904 + }, + { + "epoch": 0.08837890625, + "grad_norm": 0.19234181940555573, + "learning_rate": 0.0004958992187953256, + "loss": 1.9221, + "step": 905 + }, + { + "epoch": 0.0884765625, + "grad_norm": 0.21266022324562073, + "learning_rate": 0.0004958856927908334, + "loss": 1.9796, + "step": 906 + }, + { + "epoch": 0.08857421875, + "grad_norm": 0.22636987268924713, + "learning_rate": 0.0004958721447218174, + "loss": 1.9799, + "step": 907 + }, + { + "epoch": 0.088671875, + "grad_norm": 0.21743637323379517, + "learning_rate": 0.0004958585745896308, + "loss": 1.9156, + "step": 908 + }, + { + "epoch": 0.08876953125, + "grad_norm": 0.316840797662735, + "learning_rate": 0.0004958449823956294, + "loss": 1.9856, + "step": 909 + }, + { + "epoch": 0.0888671875, + "grad_norm": 0.2583603262901306, + "learning_rate": 0.0004958313681411706, + "loss": 1.9601, + "step": 910 + }, + { + "epoch": 0.08896484375, + "grad_norm": 0.2872125208377838, + "learning_rate": 0.0004958177318276146, + "loss": 1.9526, + "step": 911 + }, + { + "epoch": 0.0890625, + "grad_norm": 0.37896475195884705, + "learning_rate": 0.0004958040734563235, + "loss": 1.9465, + "step": 912 + }, + { + "epoch": 0.08916015625, + "grad_norm": 0.3056401014328003, + "learning_rate": 0.0004957903930286616, + "loss": 1.9344, + "step": 913 + }, + { + "epoch": 0.0892578125, + "grad_norm": 0.35092082619667053, + "learning_rate": 0.0004957766905459953, + "loss": 1.9294, + "step": 914 + }, + { + "epoch": 0.08935546875, + "grad_norm": 0.2681847810745239, + "learning_rate": 0.0004957629660096937, + "loss": 1.9436, + "step": 915 + }, + { + "epoch": 0.089453125, + "grad_norm": 0.25762611627578735, + "learning_rate": 0.0004957492194211275, + "loss": 1.931, + "step": 916 + }, + { + "epoch": 0.08955078125, + "grad_norm": 0.24634388089179993, + "learning_rate": 0.00049573545078167, + "loss": 1.9373, + "step": 917 + }, + { + "epoch": 0.0896484375, + "grad_norm": 0.27374333143234253, + "learning_rate": 0.0004957216600926965, + "loss": 2.046, + "step": 918 + }, + { + "epoch": 0.08974609375, + "grad_norm": 0.338375061750412, + "learning_rate": 0.0004957078473555845, + "loss": 2.0259, + "step": 919 + }, + { + "epoch": 0.08984375, + "grad_norm": 0.27678367495536804, + "learning_rate": 0.0004956940125717139, + "loss": 1.9947, + "step": 920 + }, + { + "epoch": 0.08994140625, + "grad_norm": 0.266079306602478, + "learning_rate": 0.0004956801557424666, + "loss": 2.0097, + "step": 921 + }, + { + "epoch": 0.0900390625, + "grad_norm": 0.28650638461112976, + "learning_rate": 0.0004956662768692268, + "loss": 1.9893, + "step": 922 + }, + { + "epoch": 0.09013671875, + "grad_norm": 0.2717742323875427, + "learning_rate": 0.0004956523759533809, + "loss": 1.976, + "step": 923 + }, + { + "epoch": 0.090234375, + "grad_norm": 0.3267741799354553, + "learning_rate": 0.0004956384529963174, + "loss": 1.9803, + "step": 924 + }, + { + "epoch": 0.09033203125, + "grad_norm": 0.32973724603652954, + "learning_rate": 0.0004956245079994272, + "loss": 1.942, + "step": 925 + }, + { + "epoch": 0.0904296875, + "grad_norm": 0.3257177770137787, + "learning_rate": 0.0004956105409641031, + "loss": 1.9584, + "step": 926 + }, + { + "epoch": 0.09052734375, + "grad_norm": 0.31881803274154663, + "learning_rate": 0.0004955965518917405, + "loss": 1.9386, + "step": 927 + }, + { + "epoch": 0.090625, + "grad_norm": 0.2814694046974182, + "learning_rate": 0.0004955825407837365, + "loss": 1.9114, + "step": 928 + }, + { + "epoch": 0.09072265625, + "grad_norm": 0.24232172966003418, + "learning_rate": 0.000495568507641491, + "loss": 1.954, + "step": 929 + }, + { + "epoch": 0.0908203125, + "grad_norm": 0.2062203288078308, + "learning_rate": 0.0004955544524664056, + "loss": 1.9682, + "step": 930 + }, + { + "epoch": 0.09091796875, + "grad_norm": 0.2777712047100067, + "learning_rate": 0.0004955403752598844, + "loss": 1.983, + "step": 931 + }, + { + "epoch": 0.091015625, + "grad_norm": 0.31253954768180847, + "learning_rate": 0.0004955262760233335, + "loss": 1.9211, + "step": 932 + }, + { + "epoch": 0.09111328125, + "grad_norm": 0.292059987783432, + "learning_rate": 0.0004955121547581613, + "loss": 1.9745, + "step": 933 + }, + { + "epoch": 0.0912109375, + "grad_norm": 0.309123694896698, + "learning_rate": 0.0004954980114657783, + "loss": 1.966, + "step": 934 + }, + { + "epoch": 0.09130859375, + "grad_norm": 0.2765352129936218, + "learning_rate": 0.0004954838461475975, + "loss": 1.9535, + "step": 935 + }, + { + "epoch": 0.09140625, + "grad_norm": 0.2576250731945038, + "learning_rate": 0.0004954696588050338, + "loss": 1.9002, + "step": 936 + }, + { + "epoch": 0.09150390625, + "grad_norm": 0.3167097866535187, + "learning_rate": 0.0004954554494395044, + "loss": 1.9249, + "step": 937 + }, + { + "epoch": 0.0916015625, + "grad_norm": 0.32505497336387634, + "learning_rate": 0.0004954412180524285, + "loss": 1.9669, + "step": 938 + }, + { + "epoch": 0.09169921875, + "grad_norm": 0.3114219307899475, + "learning_rate": 0.000495426964645228, + "loss": 1.9276, + "step": 939 + }, + { + "epoch": 0.091796875, + "grad_norm": 0.2786170542240143, + "learning_rate": 0.0004954126892193264, + "loss": 1.965, + "step": 940 + }, + { + "epoch": 0.09189453125, + "grad_norm": 0.2753596901893616, + "learning_rate": 0.00049539839177615, + "loss": 1.9925, + "step": 941 + }, + { + "epoch": 0.0919921875, + "grad_norm": 0.29401320219039917, + "learning_rate": 0.0004953840723171268, + "loss": 1.9772, + "step": 942 + }, + { + "epoch": 0.09208984375, + "grad_norm": 0.3157658278942108, + "learning_rate": 0.0004953697308436871, + "loss": 2.0019, + "step": 943 + }, + { + "epoch": 0.0921875, + "grad_norm": 0.29067566990852356, + "learning_rate": 0.0004953553673572636, + "loss": 1.9753, + "step": 944 + }, + { + "epoch": 0.09228515625, + "grad_norm": 0.28693830966949463, + "learning_rate": 0.0004953409818592911, + "loss": 1.9436, + "step": 945 + }, + { + "epoch": 0.0923828125, + "grad_norm": 0.2536534070968628, + "learning_rate": 0.0004953265743512065, + "loss": 1.9801, + "step": 946 + }, + { + "epoch": 0.09248046875, + "grad_norm": 0.3041675090789795, + "learning_rate": 0.000495312144834449, + "loss": 1.9378, + "step": 947 + }, + { + "epoch": 0.092578125, + "grad_norm": 0.3237321376800537, + "learning_rate": 0.0004952976933104602, + "loss": 1.9781, + "step": 948 + }, + { + "epoch": 0.09267578125, + "grad_norm": 0.25767290592193604, + "learning_rate": 0.0004952832197806835, + "loss": 1.9031, + "step": 949 + }, + { + "epoch": 0.0927734375, + "grad_norm": 0.25587791204452515, + "learning_rate": 0.0004952687242465645, + "loss": 1.9405, + "step": 950 + }, + { + "epoch": 0.09287109375, + "grad_norm": 0.2901047468185425, + "learning_rate": 0.0004952542067095515, + "loss": 1.9567, + "step": 951 + }, + { + "epoch": 0.09296875, + "grad_norm": 0.31517183780670166, + "learning_rate": 0.0004952396671710945, + "loss": 1.9464, + "step": 952 + }, + { + "epoch": 0.09306640625, + "grad_norm": 0.30896711349487305, + "learning_rate": 0.0004952251056326458, + "loss": 1.9096, + "step": 953 + }, + { + "epoch": 0.0931640625, + "grad_norm": 0.26281988620758057, + "learning_rate": 0.0004952105220956602, + "loss": 1.9616, + "step": 954 + }, + { + "epoch": 0.09326171875, + "grad_norm": 0.23971204459667206, + "learning_rate": 0.0004951959165615942, + "loss": 1.9928, + "step": 955 + }, + { + "epoch": 0.093359375, + "grad_norm": 0.22277623414993286, + "learning_rate": 0.000495181289031907, + "loss": 1.9652, + "step": 956 + }, + { + "epoch": 0.09345703125, + "grad_norm": 0.26211845874786377, + "learning_rate": 0.0004951666395080596, + "loss": 1.9617, + "step": 957 + }, + { + "epoch": 0.0935546875, + "grad_norm": 0.24874535202980042, + "learning_rate": 0.0004951519679915155, + "loss": 1.9509, + "step": 958 + }, + { + "epoch": 0.09365234375, + "grad_norm": 0.21750757098197937, + "learning_rate": 0.0004951372744837401, + "loss": 1.966, + "step": 959 + }, + { + "epoch": 0.09375, + "grad_norm": 0.2798830270767212, + "learning_rate": 0.0004951225589862012, + "loss": 1.9574, + "step": 960 + }, + { + "epoch": 0.09384765625, + "grad_norm": 0.33166754245758057, + "learning_rate": 0.0004951078215003688, + "loss": 1.9672, + "step": 961 + }, + { + "epoch": 0.0939453125, + "grad_norm": 0.3122573792934418, + "learning_rate": 0.0004950930620277149, + "loss": 1.9205, + "step": 962 + }, + { + "epoch": 0.09404296875, + "grad_norm": 0.233395054936409, + "learning_rate": 0.0004950782805697141, + "loss": 1.9389, + "step": 963 + }, + { + "epoch": 0.094140625, + "grad_norm": 0.2523380219936371, + "learning_rate": 0.0004950634771278427, + "loss": 1.9643, + "step": 964 + }, + { + "epoch": 0.09423828125, + "grad_norm": 0.2774633765220642, + "learning_rate": 0.0004950486517035796, + "loss": 1.9467, + "step": 965 + }, + { + "epoch": 0.0943359375, + "grad_norm": 0.2782750129699707, + "learning_rate": 0.0004950338042984056, + "loss": 1.9296, + "step": 966 + }, + { + "epoch": 0.09443359375, + "grad_norm": 0.3408234417438507, + "learning_rate": 0.0004950189349138039, + "loss": 1.9705, + "step": 967 + }, + { + "epoch": 0.09453125, + "grad_norm": 0.27266553044319153, + "learning_rate": 0.0004950040435512598, + "loss": 1.9292, + "step": 968 + }, + { + "epoch": 0.09462890625, + "grad_norm": 0.2276991307735443, + "learning_rate": 0.0004949891302122607, + "loss": 1.9793, + "step": 969 + }, + { + "epoch": 0.0947265625, + "grad_norm": 0.23914872109889984, + "learning_rate": 0.0004949741948982965, + "loss": 1.9632, + "step": 970 + }, + { + "epoch": 0.09482421875, + "grad_norm": 0.24836085736751556, + "learning_rate": 0.000494959237610859, + "loss": 1.9304, + "step": 971 + }, + { + "epoch": 0.094921875, + "grad_norm": 0.1932704746723175, + "learning_rate": 0.0004949442583514425, + "loss": 1.9483, + "step": 972 + }, + { + "epoch": 0.09501953125, + "grad_norm": 0.27378395199775696, + "learning_rate": 0.0004949292571215429, + "loss": 1.9687, + "step": 973 + }, + { + "epoch": 0.0951171875, + "grad_norm": 0.3175489008426666, + "learning_rate": 0.0004949142339226591, + "loss": 2.0036, + "step": 974 + }, + { + "epoch": 0.09521484375, + "grad_norm": 0.34563618898391724, + "learning_rate": 0.0004948991887562914, + "loss": 1.9429, + "step": 975 + }, + { + "epoch": 0.0953125, + "grad_norm": 0.2931719720363617, + "learning_rate": 0.000494884121623943, + "loss": 1.9424, + "step": 976 + }, + { + "epoch": 0.09541015625, + "grad_norm": 0.29774370789527893, + "learning_rate": 0.0004948690325271188, + "loss": 1.9297, + "step": 977 + }, + { + "epoch": 0.0955078125, + "grad_norm": 0.31513288617134094, + "learning_rate": 0.0004948539214673261, + "loss": 1.9917, + "step": 978 + }, + { + "epoch": 0.09560546875, + "grad_norm": 0.2150924950838089, + "learning_rate": 0.0004948387884460744, + "loss": 1.9529, + "step": 979 + }, + { + "epoch": 0.095703125, + "grad_norm": 0.27547362446784973, + "learning_rate": 0.0004948236334648753, + "loss": 1.9885, + "step": 980 + }, + { + "epoch": 0.09580078125, + "grad_norm": 0.3271898925304413, + "learning_rate": 0.0004948084565252427, + "loss": 1.9232, + "step": 981 + }, + { + "epoch": 0.0958984375, + "grad_norm": 0.2923772931098938, + "learning_rate": 0.0004947932576286928, + "loss": 1.9429, + "step": 982 + }, + { + "epoch": 0.09599609375, + "grad_norm": 0.2936458885669708, + "learning_rate": 0.0004947780367767434, + "loss": 1.9773, + "step": 983 + }, + { + "epoch": 0.09609375, + "grad_norm": 0.29706791043281555, + "learning_rate": 0.0004947627939709153, + "loss": 1.9284, + "step": 984 + }, + { + "epoch": 0.09619140625, + "grad_norm": 0.22721995413303375, + "learning_rate": 0.000494747529212731, + "loss": 1.9281, + "step": 985 + }, + { + "epoch": 0.0962890625, + "grad_norm": 0.27041006088256836, + "learning_rate": 0.0004947322425037152, + "loss": 1.9735, + "step": 986 + }, + { + "epoch": 0.09638671875, + "grad_norm": 0.26037418842315674, + "learning_rate": 0.0004947169338453952, + "loss": 1.9426, + "step": 987 + }, + { + "epoch": 0.096484375, + "grad_norm": 0.23383264243602753, + "learning_rate": 0.0004947016032392999, + "loss": 1.923, + "step": 988 + }, + { + "epoch": 0.09658203125, + "grad_norm": 0.29238423705101013, + "learning_rate": 0.0004946862506869609, + "loss": 1.9803, + "step": 989 + }, + { + "epoch": 0.0966796875, + "grad_norm": 0.223761647939682, + "learning_rate": 0.0004946708761899116, + "loss": 1.9377, + "step": 990 + }, + { + "epoch": 0.09677734375, + "grad_norm": 0.2669995427131653, + "learning_rate": 0.0004946554797496881, + "loss": 1.9694, + "step": 991 + }, + { + "epoch": 0.096875, + "grad_norm": 0.2847992181777954, + "learning_rate": 0.0004946400613678279, + "loss": 1.9501, + "step": 992 + }, + { + "epoch": 0.09697265625, + "grad_norm": 0.25099360942840576, + "learning_rate": 0.0004946246210458715, + "loss": 1.9707, + "step": 993 + }, + { + "epoch": 0.0970703125, + "grad_norm": 0.23306165635585785, + "learning_rate": 0.0004946091587853611, + "loss": 1.9174, + "step": 994 + }, + { + "epoch": 0.09716796875, + "grad_norm": 0.2907635569572449, + "learning_rate": 0.0004945936745878414, + "loss": 1.9521, + "step": 995 + }, + { + "epoch": 0.097265625, + "grad_norm": 0.3234463334083557, + "learning_rate": 0.0004945781684548591, + "loss": 1.9667, + "step": 996 + }, + { + "epoch": 0.09736328125, + "grad_norm": 0.36099013686180115, + "learning_rate": 0.0004945626403879629, + "loss": 1.9257, + "step": 997 + }, + { + "epoch": 0.0974609375, + "grad_norm": 0.3735397160053253, + "learning_rate": 0.0004945470903887041, + "loss": 1.9512, + "step": 998 + }, + { + "epoch": 0.09755859375, + "grad_norm": 0.3653644025325775, + "learning_rate": 0.0004945315184586359, + "loss": 1.9664, + "step": 999 + }, + { + "epoch": 0.09765625, + "grad_norm": 0.2931279242038727, + "learning_rate": 0.000494515924599314, + "loss": 1.9209, + "step": 1000 + }, + { + "epoch": 0.09775390625, + "grad_norm": 0.31677475571632385, + "learning_rate": 0.000494500308812296, + "loss": 1.9989, + "step": 1001 + }, + { + "epoch": 0.0978515625, + "grad_norm": 0.2640211582183838, + "learning_rate": 0.0004944846710991416, + "loss": 1.8792, + "step": 1002 + }, + { + "epoch": 0.09794921875, + "grad_norm": 0.3207566738128662, + "learning_rate": 0.0004944690114614132, + "loss": 1.955, + "step": 1003 + }, + { + "epoch": 0.098046875, + "grad_norm": 0.29306915402412415, + "learning_rate": 0.0004944533299006749, + "loss": 1.9264, + "step": 1004 + }, + { + "epoch": 0.09814453125, + "grad_norm": 0.2783614993095398, + "learning_rate": 0.000494437626418493, + "loss": 1.9341, + "step": 1005 + }, + { + "epoch": 0.0982421875, + "grad_norm": 0.346975713968277, + "learning_rate": 0.0004944219010164361, + "loss": 1.9599, + "step": 1006 + }, + { + "epoch": 0.09833984375, + "grad_norm": 0.2799683213233948, + "learning_rate": 0.0004944061536960755, + "loss": 1.9201, + "step": 1007 + }, + { + "epoch": 0.0984375, + "grad_norm": 0.27122223377227783, + "learning_rate": 0.0004943903844589837, + "loss": 1.9685, + "step": 1008 + }, + { + "epoch": 0.09853515625, + "grad_norm": 0.34742453694343567, + "learning_rate": 0.0004943745933067362, + "loss": 1.9533, + "step": 1009 + }, + { + "epoch": 0.0986328125, + "grad_norm": 0.2686299681663513, + "learning_rate": 0.0004943587802409103, + "loss": 1.9149, + "step": 1010 + }, + { + "epoch": 0.09873046875, + "grad_norm": 0.3223266303539276, + "learning_rate": 0.0004943429452630856, + "loss": 1.8887, + "step": 1011 + }, + { + "epoch": 0.098828125, + "grad_norm": 0.2767395079135895, + "learning_rate": 0.0004943270883748439, + "loss": 1.9756, + "step": 1012 + }, + { + "epoch": 0.09892578125, + "grad_norm": 0.2761443257331848, + "learning_rate": 0.0004943112095777691, + "loss": 1.981, + "step": 1013 + }, + { + "epoch": 0.0990234375, + "grad_norm": 0.22587864100933075, + "learning_rate": 0.0004942953088734474, + "loss": 1.939, + "step": 1014 + }, + { + "epoch": 0.09912109375, + "grad_norm": 0.2596238851547241, + "learning_rate": 0.0004942793862634671, + "loss": 1.9546, + "step": 1015 + }, + { + "epoch": 0.09921875, + "grad_norm": 0.3035464584827423, + "learning_rate": 0.0004942634417494188, + "loss": 2.0001, + "step": 1016 + }, + { + "epoch": 0.09931640625, + "grad_norm": 0.32868078351020813, + "learning_rate": 0.0004942474753328952, + "loss": 1.9248, + "step": 1017 + }, + { + "epoch": 0.0994140625, + "grad_norm": 0.34833693504333496, + "learning_rate": 0.000494231487015491, + "loss": 1.9148, + "step": 1018 + }, + { + "epoch": 0.09951171875, + "grad_norm": 0.2557753920555115, + "learning_rate": 0.0004942154767988037, + "loss": 1.9585, + "step": 1019 + }, + { + "epoch": 0.099609375, + "grad_norm": 0.25750526785850525, + "learning_rate": 0.0004941994446844321, + "loss": 2.0257, + "step": 1020 + }, + { + "epoch": 0.09970703125, + "grad_norm": 0.22088824212551117, + "learning_rate": 0.000494183390673978, + "loss": 1.9373, + "step": 1021 + }, + { + "epoch": 0.0998046875, + "grad_norm": 0.23292425274848938, + "learning_rate": 0.000494167314769045, + "loss": 1.9462, + "step": 1022 + }, + { + "epoch": 0.09990234375, + "grad_norm": 0.24202801287174225, + "learning_rate": 0.0004941512169712389, + "loss": 2.0237, + "step": 1023 + }, + { + "epoch": 0.1, + "grad_norm": 0.28507405519485474, + "learning_rate": 0.0004941350972821678, + "loss": 1.9373, + "step": 1024 + }, + { + "epoch": 0.10009765625, + "grad_norm": 0.22370004653930664, + "learning_rate": 0.0004941189557034417, + "loss": 1.9332, + "step": 1025 + }, + { + "epoch": 0.1001953125, + "grad_norm": 0.2707929313182831, + "learning_rate": 0.0004941027922366732, + "loss": 1.9908, + "step": 1026 + }, + { + "epoch": 0.10029296875, + "grad_norm": 0.29618075489997864, + "learning_rate": 0.0004940866068834769, + "loss": 1.9317, + "step": 1027 + }, + { + "epoch": 0.100390625, + "grad_norm": 0.27780136466026306, + "learning_rate": 0.0004940703996454695, + "loss": 1.9225, + "step": 1028 + }, + { + "epoch": 0.10048828125, + "grad_norm": 0.33955708146095276, + "learning_rate": 0.0004940541705242698, + "loss": 1.9232, + "step": 1029 + }, + { + "epoch": 0.1005859375, + "grad_norm": 0.3132137954235077, + "learning_rate": 0.0004940379195214992, + "loss": 1.976, + "step": 1030 + }, + { + "epoch": 0.10068359375, + "grad_norm": 0.2742595076560974, + "learning_rate": 0.000494021646638781, + "loss": 1.9138, + "step": 1031 + }, + { + "epoch": 0.10078125, + "grad_norm": 0.2921305298805237, + "learning_rate": 0.0004940053518777406, + "loss": 1.9178, + "step": 1032 + }, + { + "epoch": 0.10087890625, + "grad_norm": 0.2586468458175659, + "learning_rate": 0.0004939890352400058, + "loss": 1.9385, + "step": 1033 + }, + { + "epoch": 0.1009765625, + "grad_norm": 0.30906325578689575, + "learning_rate": 0.0004939726967272064, + "loss": 1.9368, + "step": 1034 + }, + { + "epoch": 0.10107421875, + "grad_norm": 0.3483060300350189, + "learning_rate": 0.0004939563363409745, + "loss": 1.9922, + "step": 1035 + }, + { + "epoch": 0.101171875, + "grad_norm": 0.3559792637825012, + "learning_rate": 0.0004939399540829445, + "loss": 1.9564, + "step": 1036 + }, + { + "epoch": 0.10126953125, + "grad_norm": 0.33506283164024353, + "learning_rate": 0.0004939235499547526, + "loss": 1.9784, + "step": 1037 + }, + { + "epoch": 0.1013671875, + "grad_norm": 0.23956939578056335, + "learning_rate": 0.0004939071239580377, + "loss": 1.9755, + "step": 1038 + }, + { + "epoch": 0.10146484375, + "grad_norm": 0.33348920941352844, + "learning_rate": 0.0004938906760944403, + "loss": 1.9128, + "step": 1039 + }, + { + "epoch": 0.1015625, + "grad_norm": 0.2843862771987915, + "learning_rate": 0.0004938742063656036, + "loss": 1.9297, + "step": 1040 + }, + { + "epoch": 0.10166015625, + "grad_norm": 0.2671546936035156, + "learning_rate": 0.0004938577147731727, + "loss": 1.9391, + "step": 1041 + }, + { + "epoch": 0.1017578125, + "grad_norm": 0.2899993360042572, + "learning_rate": 0.000493841201318795, + "loss": 1.9198, + "step": 1042 + }, + { + "epoch": 0.10185546875, + "grad_norm": 0.20722518861293793, + "learning_rate": 0.0004938246660041201, + "loss": 1.9295, + "step": 1043 + }, + { + "epoch": 0.101953125, + "grad_norm": 0.22667360305786133, + "learning_rate": 0.0004938081088307997, + "loss": 1.9371, + "step": 1044 + }, + { + "epoch": 0.10205078125, + "grad_norm": 0.275579035282135, + "learning_rate": 0.0004937915298004877, + "loss": 1.9383, + "step": 1045 + }, + { + "epoch": 0.1021484375, + "grad_norm": 0.2900732159614563, + "learning_rate": 0.0004937749289148401, + "loss": 1.9805, + "step": 1046 + }, + { + "epoch": 0.10224609375, + "grad_norm": 0.26479572057724, + "learning_rate": 0.0004937583061755153, + "loss": 1.911, + "step": 1047 + }, + { + "epoch": 0.10234375, + "grad_norm": 0.2967575490474701, + "learning_rate": 0.0004937416615841739, + "loss": 1.9654, + "step": 1048 + }, + { + "epoch": 0.10244140625, + "grad_norm": 0.2840502858161926, + "learning_rate": 0.0004937249951424782, + "loss": 1.9332, + "step": 1049 + }, + { + "epoch": 0.1025390625, + "grad_norm": 0.2664666175842285, + "learning_rate": 0.0004937083068520931, + "loss": 1.973, + "step": 1050 + }, + { + "epoch": 0.10263671875, + "grad_norm": 0.2535867989063263, + "learning_rate": 0.000493691596714686, + "loss": 1.9421, + "step": 1051 + }, + { + "epoch": 0.102734375, + "grad_norm": 0.3005189895629883, + "learning_rate": 0.0004936748647319256, + "loss": 1.9434, + "step": 1052 + }, + { + "epoch": 0.10283203125, + "grad_norm": 0.2663765549659729, + "learning_rate": 0.0004936581109054837, + "loss": 1.9947, + "step": 1053 + }, + { + "epoch": 0.1029296875, + "grad_norm": 0.29048940539360046, + "learning_rate": 0.0004936413352370335, + "loss": 1.9737, + "step": 1054 + }, + { + "epoch": 0.10302734375, + "grad_norm": 0.2594558596611023, + "learning_rate": 0.000493624537728251, + "loss": 1.9601, + "step": 1055 + }, + { + "epoch": 0.103125, + "grad_norm": 0.27011555433273315, + "learning_rate": 0.000493607718380814, + "loss": 1.9459, + "step": 1056 + }, + { + "epoch": 0.10322265625, + "grad_norm": 0.3043757677078247, + "learning_rate": 0.0004935908771964026, + "loss": 1.9494, + "step": 1057 + }, + { + "epoch": 0.1033203125, + "grad_norm": 0.26345208287239075, + "learning_rate": 0.0004935740141766991, + "loss": 1.896, + "step": 1058 + }, + { + "epoch": 0.10341796875, + "grad_norm": 0.20400577783584595, + "learning_rate": 0.000493557129323388, + "loss": 1.9276, + "step": 1059 + }, + { + "epoch": 0.103515625, + "grad_norm": 0.2836177349090576, + "learning_rate": 0.0004935402226381558, + "loss": 1.921, + "step": 1060 + }, + { + "epoch": 0.10361328125, + "grad_norm": 0.2904714047908783, + "learning_rate": 0.0004935232941226916, + "loss": 1.9631, + "step": 1061 + }, + { + "epoch": 0.1037109375, + "grad_norm": 0.25980454683303833, + "learning_rate": 0.0004935063437786863, + "loss": 1.9757, + "step": 1062 + }, + { + "epoch": 0.10380859375, + "grad_norm": 0.27394455671310425, + "learning_rate": 0.0004934893716078329, + "loss": 1.9647, + "step": 1063 + }, + { + "epoch": 0.10390625, + "grad_norm": 0.23419839143753052, + "learning_rate": 0.000493472377611827, + "loss": 1.962, + "step": 1064 + }, + { + "epoch": 0.10400390625, + "grad_norm": 0.2323913723230362, + "learning_rate": 0.0004934553617923661, + "loss": 1.9759, + "step": 1065 + }, + { + "epoch": 0.1041015625, + "grad_norm": 0.2700064778327942, + "learning_rate": 0.0004934383241511498, + "loss": 1.9206, + "step": 1066 + }, + { + "epoch": 0.10419921875, + "grad_norm": 0.3530958294868469, + "learning_rate": 0.0004934212646898802, + "loss": 1.979, + "step": 1067 + }, + { + "epoch": 0.104296875, + "grad_norm": 0.30405712127685547, + "learning_rate": 0.0004934041834102612, + "loss": 1.951, + "step": 1068 + }, + { + "epoch": 0.10439453125, + "grad_norm": 0.2923583984375, + "learning_rate": 0.0004933870803139992, + "loss": 2.0158, + "step": 1069 + }, + { + "epoch": 0.1044921875, + "grad_norm": 0.23782993853092194, + "learning_rate": 0.0004933699554028027, + "loss": 1.896, + "step": 1070 + }, + { + "epoch": 0.10458984375, + "grad_norm": 0.3045034110546112, + "learning_rate": 0.0004933528086783821, + "loss": 1.9555, + "step": 1071 + }, + { + "epoch": 0.1046875, + "grad_norm": 0.40851131081581116, + "learning_rate": 0.0004933356401424505, + "loss": 1.9108, + "step": 1072 + }, + { + "epoch": 0.10478515625, + "grad_norm": 0.39787429571151733, + "learning_rate": 0.0004933184497967226, + "loss": 1.9472, + "step": 1073 + }, + { + "epoch": 0.1048828125, + "grad_norm": 0.24665354192256927, + "learning_rate": 0.0004933012376429158, + "loss": 1.9617, + "step": 1074 + }, + { + "epoch": 0.10498046875, + "grad_norm": 0.28025394678115845, + "learning_rate": 0.0004932840036827492, + "loss": 1.898, + "step": 1075 + }, + { + "epoch": 0.105078125, + "grad_norm": 0.3374732434749603, + "learning_rate": 0.0004932667479179445, + "loss": 1.9323, + "step": 1076 + }, + { + "epoch": 0.10517578125, + "grad_norm": 0.2941461503505707, + "learning_rate": 0.0004932494703502254, + "loss": 1.964, + "step": 1077 + }, + { + "epoch": 0.1052734375, + "grad_norm": 0.2827446758747101, + "learning_rate": 0.0004932321709813177, + "loss": 1.9554, + "step": 1078 + }, + { + "epoch": 0.10537109375, + "grad_norm": 0.32132452726364136, + "learning_rate": 0.0004932148498129495, + "loss": 1.9333, + "step": 1079 + }, + { + "epoch": 0.10546875, + "grad_norm": 0.2732277810573578, + "learning_rate": 0.000493197506846851, + "loss": 1.932, + "step": 1080 + }, + { + "epoch": 0.10556640625, + "grad_norm": 0.2594139575958252, + "learning_rate": 0.0004931801420847545, + "loss": 1.9447, + "step": 1081 + }, + { + "epoch": 0.1056640625, + "grad_norm": 0.24399201571941376, + "learning_rate": 0.0004931627555283948, + "loss": 1.9163, + "step": 1082 + }, + { + "epoch": 0.10576171875, + "grad_norm": 0.22696708142757416, + "learning_rate": 0.0004931453471795087, + "loss": 1.9734, + "step": 1083 + }, + { + "epoch": 0.105859375, + "grad_norm": 0.25717827677726746, + "learning_rate": 0.0004931279170398349, + "loss": 1.9767, + "step": 1084 + }, + { + "epoch": 0.10595703125, + "grad_norm": 0.23899690806865692, + "learning_rate": 0.0004931104651111146, + "loss": 1.9519, + "step": 1085 + }, + { + "epoch": 0.1060546875, + "grad_norm": 0.27689501643180847, + "learning_rate": 0.0004930929913950911, + "loss": 1.9582, + "step": 1086 + }, + { + "epoch": 0.10615234375, + "grad_norm": 0.23959669470787048, + "learning_rate": 0.00049307549589351, + "loss": 1.9369, + "step": 1087 + }, + { + "epoch": 0.10625, + "grad_norm": 0.25673651695251465, + "learning_rate": 0.0004930579786081188, + "loss": 1.9711, + "step": 1088 + }, + { + "epoch": 0.10634765625, + "grad_norm": 0.30868396162986755, + "learning_rate": 0.0004930404395406674, + "loss": 1.9539, + "step": 1089 + }, + { + "epoch": 0.1064453125, + "grad_norm": 0.2462042272090912, + "learning_rate": 0.0004930228786929076, + "loss": 1.9061, + "step": 1090 + }, + { + "epoch": 0.10654296875, + "grad_norm": 0.24275164306163788, + "learning_rate": 0.0004930052960665938, + "loss": 1.8992, + "step": 1091 + }, + { + "epoch": 0.106640625, + "grad_norm": 0.27242550253868103, + "learning_rate": 0.0004929876916634823, + "loss": 1.8997, + "step": 1092 + }, + { + "epoch": 0.10673828125, + "grad_norm": 0.2655372619628906, + "learning_rate": 0.0004929700654853316, + "loss": 1.8905, + "step": 1093 + }, + { + "epoch": 0.1068359375, + "grad_norm": 0.27779421210289, + "learning_rate": 0.0004929524175339024, + "loss": 1.9485, + "step": 1094 + }, + { + "epoch": 0.10693359375, + "grad_norm": 0.3308311402797699, + "learning_rate": 0.0004929347478109576, + "loss": 1.896, + "step": 1095 + }, + { + "epoch": 0.10703125, + "grad_norm": 0.2988120913505554, + "learning_rate": 0.0004929170563182621, + "loss": 1.9508, + "step": 1096 + }, + { + "epoch": 0.10712890625, + "grad_norm": 0.24604608118534088, + "learning_rate": 0.0004928993430575834, + "loss": 1.9912, + "step": 1097 + }, + { + "epoch": 0.1072265625, + "grad_norm": 0.31434887647628784, + "learning_rate": 0.0004928816080306906, + "loss": 1.9592, + "step": 1098 + }, + { + "epoch": 0.10732421875, + "grad_norm": 0.21227046847343445, + "learning_rate": 0.0004928638512393555, + "loss": 1.9523, + "step": 1099 + }, + { + "epoch": 0.107421875, + "grad_norm": 0.26643091440200806, + "learning_rate": 0.0004928460726853518, + "loss": 1.9506, + "step": 1100 + }, + { + "epoch": 0.10751953125, + "grad_norm": 0.2661186754703522, + "learning_rate": 0.0004928282723704554, + "loss": 1.9425, + "step": 1101 + }, + { + "epoch": 0.1076171875, + "grad_norm": 0.25391289591789246, + "learning_rate": 0.0004928104502964442, + "loss": 1.9551, + "step": 1102 + }, + { + "epoch": 0.10771484375, + "grad_norm": 0.276909202337265, + "learning_rate": 0.0004927926064650988, + "loss": 1.9416, + "step": 1103 + }, + { + "epoch": 0.1078125, + "grad_norm": 0.25135815143585205, + "learning_rate": 0.0004927747408782014, + "loss": 1.9197, + "step": 1104 + }, + { + "epoch": 0.10791015625, + "grad_norm": 0.2529551684856415, + "learning_rate": 0.0004927568535375368, + "loss": 1.9561, + "step": 1105 + }, + { + "epoch": 0.1080078125, + "grad_norm": 0.2697699964046478, + "learning_rate": 0.0004927389444448915, + "loss": 1.9369, + "step": 1106 + }, + { + "epoch": 0.10810546875, + "grad_norm": 0.2805427014827728, + "learning_rate": 0.0004927210136020548, + "loss": 1.9535, + "step": 1107 + }, + { + "epoch": 0.108203125, + "grad_norm": 0.33747565746307373, + "learning_rate": 0.0004927030610108176, + "loss": 1.9582, + "step": 1108 + }, + { + "epoch": 0.10830078125, + "grad_norm": 0.30215132236480713, + "learning_rate": 0.0004926850866729734, + "loss": 1.9245, + "step": 1109 + }, + { + "epoch": 0.1083984375, + "grad_norm": 0.3483729064464569, + "learning_rate": 0.0004926670905903174, + "loss": 1.9959, + "step": 1110 + }, + { + "epoch": 0.10849609375, + "grad_norm": 0.23263247311115265, + "learning_rate": 0.0004926490727646476, + "loss": 1.9224, + "step": 1111 + }, + { + "epoch": 0.10859375, + "grad_norm": 0.3023853898048401, + "learning_rate": 0.0004926310331977633, + "loss": 1.9493, + "step": 1112 + }, + { + "epoch": 0.10869140625, + "grad_norm": 0.3592170774936676, + "learning_rate": 0.0004926129718914672, + "loss": 1.9082, + "step": 1113 + }, + { + "epoch": 0.1087890625, + "grad_norm": 0.29466864466667175, + "learning_rate": 0.0004925948888475628, + "loss": 2.016, + "step": 1114 + }, + { + "epoch": 0.10888671875, + "grad_norm": 0.3136419355869293, + "learning_rate": 0.0004925767840678569, + "loss": 1.9592, + "step": 1115 + }, + { + "epoch": 0.108984375, + "grad_norm": 0.21686695516109467, + "learning_rate": 0.0004925586575541578, + "loss": 1.9081, + "step": 1116 + }, + { + "epoch": 0.10908203125, + "grad_norm": 0.2520851790904999, + "learning_rate": 0.0004925405093082761, + "loss": 1.9418, + "step": 1117 + }, + { + "epoch": 0.1091796875, + "grad_norm": 0.28278669714927673, + "learning_rate": 0.0004925223393320247, + "loss": 1.9815, + "step": 1118 + }, + { + "epoch": 0.10927734375, + "grad_norm": 0.2773968577384949, + "learning_rate": 0.0004925041476272189, + "loss": 1.8888, + "step": 1119 + }, + { + "epoch": 0.109375, + "grad_norm": 0.24399970471858978, + "learning_rate": 0.0004924859341956755, + "loss": 1.9548, + "step": 1120 + }, + { + "epoch": 0.10947265625, + "grad_norm": 0.26073598861694336, + "learning_rate": 0.0004924676990392141, + "loss": 1.9112, + "step": 1121 + }, + { + "epoch": 0.1095703125, + "grad_norm": 0.24541547894477844, + "learning_rate": 0.0004924494421596562, + "loss": 1.914, + "step": 1122 + }, + { + "epoch": 0.10966796875, + "grad_norm": 0.21719145774841309, + "learning_rate": 0.0004924311635588254, + "loss": 1.9204, + "step": 1123 + }, + { + "epoch": 0.109765625, + "grad_norm": 0.295460045337677, + "learning_rate": 0.0004924128632385476, + "loss": 1.9545, + "step": 1124 + }, + { + "epoch": 0.10986328125, + "grad_norm": 0.3058072626590729, + "learning_rate": 0.0004923945412006509, + "loss": 1.9048, + "step": 1125 + }, + { + "epoch": 0.1099609375, + "grad_norm": 0.19512003660202026, + "learning_rate": 0.0004923761974469655, + "loss": 1.9581, + "step": 1126 + }, + { + "epoch": 0.11005859375, + "grad_norm": 0.21684713661670685, + "learning_rate": 0.0004923578319793237, + "loss": 1.8765, + "step": 1127 + }, + { + "epoch": 0.11015625, + "grad_norm": 0.2229132503271103, + "learning_rate": 0.0004923394447995601, + "loss": 1.8599, + "step": 1128 + }, + { + "epoch": 0.11025390625, + "grad_norm": 0.20479273796081543, + "learning_rate": 0.0004923210359095115, + "loss": 1.938, + "step": 1129 + }, + { + "epoch": 0.1103515625, + "grad_norm": 0.2477833479642868, + "learning_rate": 0.0004923026053110167, + "loss": 1.9655, + "step": 1130 + }, + { + "epoch": 0.11044921875, + "grad_norm": 0.23927846550941467, + "learning_rate": 0.0004922841530059167, + "loss": 1.9461, + "step": 1131 + }, + { + "epoch": 0.110546875, + "grad_norm": 0.24189342558383942, + "learning_rate": 0.0004922656789960547, + "loss": 1.9338, + "step": 1132 + }, + { + "epoch": 0.11064453125, + "grad_norm": 0.20208942890167236, + "learning_rate": 0.0004922471832832763, + "loss": 1.973, + "step": 1133 + }, + { + "epoch": 0.1107421875, + "grad_norm": 0.20967181026935577, + "learning_rate": 0.0004922286658694289, + "loss": 1.9391, + "step": 1134 + }, + { + "epoch": 0.11083984375, + "grad_norm": 0.2538667917251587, + "learning_rate": 0.0004922101267563622, + "loss": 1.9159, + "step": 1135 + }, + { + "epoch": 0.1109375, + "grad_norm": 0.2945930063724518, + "learning_rate": 0.0004921915659459281, + "loss": 1.915, + "step": 1136 + }, + { + "epoch": 0.11103515625, + "grad_norm": 0.3631344437599182, + "learning_rate": 0.0004921729834399808, + "loss": 1.9926, + "step": 1137 + }, + { + "epoch": 0.1111328125, + "grad_norm": 0.32187792658805847, + "learning_rate": 0.0004921543792403765, + "loss": 1.9137, + "step": 1138 + }, + { + "epoch": 0.11123046875, + "grad_norm": 0.276929646730423, + "learning_rate": 0.0004921357533489734, + "loss": 1.9427, + "step": 1139 + }, + { + "epoch": 0.111328125, + "grad_norm": 0.20455695688724518, + "learning_rate": 0.0004921171057676323, + "loss": 1.9118, + "step": 1140 + }, + { + "epoch": 0.11142578125, + "grad_norm": 0.2824450731277466, + "learning_rate": 0.0004920984364982157, + "loss": 1.9471, + "step": 1141 + }, + { + "epoch": 0.1115234375, + "grad_norm": 0.2931687831878662, + "learning_rate": 0.0004920797455425887, + "loss": 1.9546, + "step": 1142 + }, + { + "epoch": 0.11162109375, + "grad_norm": 0.31234028935432434, + "learning_rate": 0.0004920610329026181, + "loss": 1.9466, + "step": 1143 + }, + { + "epoch": 0.11171875, + "grad_norm": 0.39190223813056946, + "learning_rate": 0.0004920422985801735, + "loss": 1.9394, + "step": 1144 + }, + { + "epoch": 0.11181640625, + "grad_norm": 0.26808232069015503, + "learning_rate": 0.0004920235425771261, + "loss": 1.9278, + "step": 1145 + }, + { + "epoch": 0.1119140625, + "grad_norm": 0.22590310871601105, + "learning_rate": 0.0004920047648953493, + "loss": 1.9467, + "step": 1146 + }, + { + "epoch": 0.11201171875, + "grad_norm": 0.2531168758869171, + "learning_rate": 0.0004919859655367192, + "loss": 1.958, + "step": 1147 + }, + { + "epoch": 0.112109375, + "grad_norm": 0.207502081990242, + "learning_rate": 0.0004919671445031132, + "loss": 2.0148, + "step": 1148 + }, + { + "epoch": 0.11220703125, + "grad_norm": 0.23596623539924622, + "learning_rate": 0.0004919483017964118, + "loss": 1.9617, + "step": 1149 + }, + { + "epoch": 0.1123046875, + "grad_norm": 0.25810694694519043, + "learning_rate": 0.000491929437418497, + "loss": 2.0262, + "step": 1150 + }, + { + "epoch": 0.11240234375, + "grad_norm": 0.27257680892944336, + "learning_rate": 0.0004919105513712533, + "loss": 1.9285, + "step": 1151 + }, + { + "epoch": 0.1125, + "grad_norm": 0.2861190736293793, + "learning_rate": 0.000491891643656567, + "loss": 1.965, + "step": 1152 + }, + { + "epoch": 0.11259765625, + "grad_norm": 0.2790696620941162, + "learning_rate": 0.0004918727142763271, + "loss": 1.9572, + "step": 1153 + }, + { + "epoch": 0.1126953125, + "grad_norm": 0.35226336121559143, + "learning_rate": 0.0004918537632324243, + "loss": 1.9338, + "step": 1154 + }, + { + "epoch": 0.11279296875, + "grad_norm": 0.2735171616077423, + "learning_rate": 0.0004918347905267517, + "loss": 1.9304, + "step": 1155 + }, + { + "epoch": 0.112890625, + "grad_norm": 0.2629276514053345, + "learning_rate": 0.0004918157961612046, + "loss": 1.9383, + "step": 1156 + }, + { + "epoch": 0.11298828125, + "grad_norm": 0.3583827614784241, + "learning_rate": 0.0004917967801376802, + "loss": 1.9448, + "step": 1157 + }, + { + "epoch": 0.1130859375, + "grad_norm": 0.2546387016773224, + "learning_rate": 0.0004917777424580781, + "loss": 1.9506, + "step": 1158 + }, + { + "epoch": 0.11318359375, + "grad_norm": 0.35346466302871704, + "learning_rate": 0.0004917586831243001, + "loss": 1.9207, + "step": 1159 + }, + { + "epoch": 0.11328125, + "grad_norm": 0.2943922281265259, + "learning_rate": 0.0004917396021382499, + "loss": 1.961, + "step": 1160 + }, + { + "epoch": 0.11337890625, + "grad_norm": 0.24972833693027496, + "learning_rate": 0.0004917204995018336, + "loss": 1.9271, + "step": 1161 + }, + { + "epoch": 0.1134765625, + "grad_norm": 0.31200850009918213, + "learning_rate": 0.0004917013752169594, + "loss": 1.9444, + "step": 1162 + }, + { + "epoch": 0.11357421875, + "grad_norm": 0.3580279052257538, + "learning_rate": 0.0004916822292855375, + "loss": 1.9132, + "step": 1163 + }, + { + "epoch": 0.113671875, + "grad_norm": 0.31269049644470215, + "learning_rate": 0.0004916630617094807, + "loss": 1.9283, + "step": 1164 + }, + { + "epoch": 0.11376953125, + "grad_norm": 0.25159773230552673, + "learning_rate": 0.0004916438724907033, + "loss": 1.9018, + "step": 1165 + }, + { + "epoch": 0.1138671875, + "grad_norm": 0.22316916286945343, + "learning_rate": 0.0004916246616311225, + "loss": 1.8821, + "step": 1166 + }, + { + "epoch": 0.11396484375, + "grad_norm": 0.31150946021080017, + "learning_rate": 0.0004916054291326571, + "loss": 1.9565, + "step": 1167 + }, + { + "epoch": 0.1140625, + "grad_norm": 0.29112792015075684, + "learning_rate": 0.0004915861749972281, + "loss": 1.9582, + "step": 1168 + }, + { + "epoch": 0.11416015625, + "grad_norm": 0.26860764622688293, + "learning_rate": 0.0004915668992267593, + "loss": 1.9299, + "step": 1169 + }, + { + "epoch": 0.1142578125, + "grad_norm": 0.24470628798007965, + "learning_rate": 0.0004915476018231757, + "loss": 1.9131, + "step": 1170 + }, + { + "epoch": 0.11435546875, + "grad_norm": 0.26279789209365845, + "learning_rate": 0.0004915282827884051, + "loss": 1.9055, + "step": 1171 + }, + { + "epoch": 0.114453125, + "grad_norm": 0.21272587776184082, + "learning_rate": 0.0004915089421243773, + "loss": 1.9694, + "step": 1172 + }, + { + "epoch": 0.11455078125, + "grad_norm": 0.21711157262325287, + "learning_rate": 0.0004914895798330244, + "loss": 1.9644, + "step": 1173 + }, + { + "epoch": 0.1146484375, + "grad_norm": 0.21599583327770233, + "learning_rate": 0.0004914701959162803, + "loss": 1.9528, + "step": 1174 + }, + { + "epoch": 0.11474609375, + "grad_norm": 0.22919407486915588, + "learning_rate": 0.0004914507903760814, + "loss": 1.9353, + "step": 1175 + }, + { + "epoch": 0.11484375, + "grad_norm": 0.22974969446659088, + "learning_rate": 0.0004914313632143661, + "loss": 1.9148, + "step": 1176 + }, + { + "epoch": 0.11494140625, + "grad_norm": 0.2370641529560089, + "learning_rate": 0.000491411914433075, + "loss": 1.9056, + "step": 1177 + }, + { + "epoch": 0.1150390625, + "grad_norm": 0.27402329444885254, + "learning_rate": 0.000491392444034151, + "loss": 1.9483, + "step": 1178 + }, + { + "epoch": 0.11513671875, + "grad_norm": 0.2953847050666809, + "learning_rate": 0.0004913729520195388, + "loss": 1.9553, + "step": 1179 + }, + { + "epoch": 0.115234375, + "grad_norm": 0.26085543632507324, + "learning_rate": 0.0004913534383911857, + "loss": 1.9307, + "step": 1180 + }, + { + "epoch": 0.11533203125, + "grad_norm": 0.2782340943813324, + "learning_rate": 0.0004913339031510408, + "loss": 1.9733, + "step": 1181 + }, + { + "epoch": 0.1154296875, + "grad_norm": 0.2568228244781494, + "learning_rate": 0.0004913143463010554, + "loss": 1.9402, + "step": 1182 + }, + { + "epoch": 0.11552734375, + "grad_norm": 0.2584892809391022, + "learning_rate": 0.0004912947678431833, + "loss": 1.9254, + "step": 1183 + }, + { + "epoch": 0.115625, + "grad_norm": 0.2621140778064728, + "learning_rate": 0.00049127516777938, + "loss": 1.9323, + "step": 1184 + }, + { + "epoch": 0.11572265625, + "grad_norm": 0.28239843249320984, + "learning_rate": 0.0004912555461116035, + "loss": 1.9447, + "step": 1185 + }, + { + "epoch": 0.1158203125, + "grad_norm": 0.3147210478782654, + "learning_rate": 0.0004912359028418138, + "loss": 1.8867, + "step": 1186 + }, + { + "epoch": 0.11591796875, + "grad_norm": 0.25541990995407104, + "learning_rate": 0.0004912162379719731, + "loss": 1.871, + "step": 1187 + }, + { + "epoch": 0.116015625, + "grad_norm": 0.24328139424324036, + "learning_rate": 0.0004911965515040457, + "loss": 1.9316, + "step": 1188 + }, + { + "epoch": 0.11611328125, + "grad_norm": 0.20230042934417725, + "learning_rate": 0.0004911768434399981, + "loss": 1.9298, + "step": 1189 + }, + { + "epoch": 0.1162109375, + "grad_norm": 0.32389941811561584, + "learning_rate": 0.0004911571137817991, + "loss": 1.8941, + "step": 1190 + }, + { + "epoch": 0.11630859375, + "grad_norm": 0.3173327147960663, + "learning_rate": 0.0004911373625314194, + "loss": 1.9222, + "step": 1191 + }, + { + "epoch": 0.11640625, + "grad_norm": 0.32637453079223633, + "learning_rate": 0.0004911175896908319, + "loss": 1.9273, + "step": 1192 + }, + { + "epoch": 0.11650390625, + "grad_norm": 0.307746022939682, + "learning_rate": 0.0004910977952620119, + "loss": 1.9352, + "step": 1193 + }, + { + "epoch": 0.1166015625, + "grad_norm": 0.21165402233600616, + "learning_rate": 0.0004910779792469365, + "loss": 1.8867, + "step": 1194 + }, + { + "epoch": 0.11669921875, + "grad_norm": 0.3059009909629822, + "learning_rate": 0.0004910581416475854, + "loss": 1.9187, + "step": 1195 + }, + { + "epoch": 0.116796875, + "grad_norm": 0.2690452039241791, + "learning_rate": 0.00049103828246594, + "loss": 1.9723, + "step": 1196 + }, + { + "epoch": 0.11689453125, + "grad_norm": 0.25487321615219116, + "learning_rate": 0.0004910184017039841, + "loss": 1.9525, + "step": 1197 + }, + { + "epoch": 0.1169921875, + "grad_norm": 0.33227404952049255, + "learning_rate": 0.0004909984993637036, + "loss": 1.9062, + "step": 1198 + }, + { + "epoch": 0.11708984375, + "grad_norm": 0.32993388175964355, + "learning_rate": 0.0004909785754470866, + "loss": 1.9327, + "step": 1199 + }, + { + "epoch": 0.1171875, + "grad_norm": 0.28574901819229126, + "learning_rate": 0.0004909586299561233, + "loss": 1.9481, + "step": 1200 + }, + { + "epoch": 0.11728515625, + "grad_norm": 0.27981141209602356, + "learning_rate": 0.0004909386628928061, + "loss": 1.9284, + "step": 1201 + }, + { + "epoch": 0.1173828125, + "grad_norm": 0.26131895184516907, + "learning_rate": 0.0004909186742591296, + "loss": 1.9276, + "step": 1202 + }, + { + "epoch": 0.11748046875, + "grad_norm": 0.3690519630908966, + "learning_rate": 0.0004908986640570903, + "loss": 1.8733, + "step": 1203 + }, + { + "epoch": 0.117578125, + "grad_norm": 0.38982152938842773, + "learning_rate": 0.000490878632288687, + "loss": 1.9204, + "step": 1204 + }, + { + "epoch": 0.11767578125, + "grad_norm": 0.2588847875595093, + "learning_rate": 0.000490858578955921, + "loss": 1.8941, + "step": 1205 + }, + { + "epoch": 0.1177734375, + "grad_norm": 0.3161073625087738, + "learning_rate": 0.0004908385040607954, + "loss": 1.9077, + "step": 1206 + }, + { + "epoch": 0.11787109375, + "grad_norm": 0.26444011926651, + "learning_rate": 0.0004908184076053153, + "loss": 1.9471, + "step": 1207 + }, + { + "epoch": 0.11796875, + "grad_norm": 0.28895479440689087, + "learning_rate": 0.0004907982895914883, + "loss": 1.9159, + "step": 1208 + }, + { + "epoch": 0.11806640625, + "grad_norm": 0.264992356300354, + "learning_rate": 0.000490778150021324, + "loss": 1.9383, + "step": 1209 + }, + { + "epoch": 0.1181640625, + "grad_norm": 0.25046393275260925, + "learning_rate": 0.0004907579888968341, + "loss": 1.9756, + "step": 1210 + }, + { + "epoch": 0.11826171875, + "grad_norm": 0.2875134348869324, + "learning_rate": 0.0004907378062200325, + "loss": 1.9368, + "step": 1211 + }, + { + "epoch": 0.118359375, + "grad_norm": 0.24127423763275146, + "learning_rate": 0.0004907176019929353, + "loss": 1.9399, + "step": 1212 + }, + { + "epoch": 0.11845703125, + "grad_norm": 0.25520575046539307, + "learning_rate": 0.0004906973762175608, + "loss": 1.9662, + "step": 1213 + }, + { + "epoch": 0.1185546875, + "grad_norm": 0.23595774173736572, + "learning_rate": 0.0004906771288959293, + "loss": 1.9302, + "step": 1214 + }, + { + "epoch": 0.11865234375, + "grad_norm": 0.2919490337371826, + "learning_rate": 0.0004906568600300635, + "loss": 1.9421, + "step": 1215 + }, + { + "epoch": 0.11875, + "grad_norm": 0.23202334344387054, + "learning_rate": 0.0004906365696219877, + "loss": 1.8934, + "step": 1216 + }, + { + "epoch": 0.11884765625, + "grad_norm": 0.2639678716659546, + "learning_rate": 0.0004906162576737292, + "loss": 1.9272, + "step": 1217 + }, + { + "epoch": 0.1189453125, + "grad_norm": 0.24962055683135986, + "learning_rate": 0.0004905959241873165, + "loss": 1.9605, + "step": 1218 + }, + { + "epoch": 0.11904296875, + "grad_norm": 0.2519899308681488, + "learning_rate": 0.0004905755691647812, + "loss": 1.9757, + "step": 1219 + }, + { + "epoch": 0.119140625, + "grad_norm": 0.2044687420129776, + "learning_rate": 0.0004905551926081563, + "loss": 1.9817, + "step": 1220 + }, + { + "epoch": 0.11923828125, + "grad_norm": 0.21942391991615295, + "learning_rate": 0.0004905347945194772, + "loss": 1.9705, + "step": 1221 + }, + { + "epoch": 0.1193359375, + "grad_norm": 0.26614969968795776, + "learning_rate": 0.0004905143749007817, + "loss": 1.9375, + "step": 1222 + }, + { + "epoch": 0.11943359375, + "grad_norm": 0.23212748765945435, + "learning_rate": 0.0004904939337541093, + "loss": 1.9728, + "step": 1223 + }, + { + "epoch": 0.11953125, + "grad_norm": 0.28004127740859985, + "learning_rate": 0.0004904734710815022, + "loss": 1.9082, + "step": 1224 + }, + { + "epoch": 0.11962890625, + "grad_norm": 0.3012614846229553, + "learning_rate": 0.0004904529868850042, + "loss": 1.9347, + "step": 1225 + }, + { + "epoch": 0.1197265625, + "grad_norm": 0.2914253771305084, + "learning_rate": 0.0004904324811666615, + "loss": 1.9435, + "step": 1226 + }, + { + "epoch": 0.11982421875, + "grad_norm": 0.2315436154603958, + "learning_rate": 0.0004904119539285225, + "loss": 1.9503, + "step": 1227 + }, + { + "epoch": 0.119921875, + "grad_norm": 0.2043173611164093, + "learning_rate": 0.0004903914051726377, + "loss": 1.937, + "step": 1228 + }, + { + "epoch": 0.12001953125, + "grad_norm": 0.26184800267219543, + "learning_rate": 0.0004903708349010598, + "loss": 1.9459, + "step": 1229 + }, + { + "epoch": 0.1201171875, + "grad_norm": 0.24898020923137665, + "learning_rate": 0.0004903502431158434, + "loss": 1.935, + "step": 1230 + }, + { + "epoch": 0.12021484375, + "grad_norm": 0.29706010222435, + "learning_rate": 0.0004903296298190456, + "loss": 1.9741, + "step": 1231 + }, + { + "epoch": 0.1203125, + "grad_norm": 0.3235405385494232, + "learning_rate": 0.0004903089950127254, + "loss": 1.9319, + "step": 1232 + }, + { + "epoch": 0.12041015625, + "grad_norm": 0.25629666447639465, + "learning_rate": 0.0004902883386989441, + "loss": 1.9533, + "step": 1233 + }, + { + "epoch": 0.1205078125, + "grad_norm": 0.23315580189228058, + "learning_rate": 0.0004902676608797649, + "loss": 1.9033, + "step": 1234 + }, + { + "epoch": 0.12060546875, + "grad_norm": 0.2336481362581253, + "learning_rate": 0.0004902469615572537, + "loss": 1.9123, + "step": 1235 + }, + { + "epoch": 0.120703125, + "grad_norm": 0.27162161469459534, + "learning_rate": 0.0004902262407334779, + "loss": 1.9253, + "step": 1236 + }, + { + "epoch": 0.12080078125, + "grad_norm": 0.2913827896118164, + "learning_rate": 0.0004902054984105073, + "loss": 1.9526, + "step": 1237 + }, + { + "epoch": 0.1208984375, + "grad_norm": 0.30284538865089417, + "learning_rate": 0.0004901847345904141, + "loss": 1.9588, + "step": 1238 + }, + { + "epoch": 0.12099609375, + "grad_norm": 0.2579716742038727, + "learning_rate": 0.000490163949275272, + "loss": 1.892, + "step": 1239 + }, + { + "epoch": 0.12109375, + "grad_norm": 0.21740810573101044, + "learning_rate": 0.0004901431424671577, + "loss": 1.9584, + "step": 1240 + }, + { + "epoch": 0.12119140625, + "grad_norm": 0.3371858298778534, + "learning_rate": 0.0004901223141681496, + "loss": 1.9415, + "step": 1241 + }, + { + "epoch": 0.1212890625, + "grad_norm": 0.34596753120422363, + "learning_rate": 0.000490101464380328, + "loss": 1.9384, + "step": 1242 + }, + { + "epoch": 0.12138671875, + "grad_norm": 0.2591091990470886, + "learning_rate": 0.0004900805931057757, + "loss": 1.9433, + "step": 1243 + }, + { + "epoch": 0.121484375, + "grad_norm": 0.27759140729904175, + "learning_rate": 0.0004900597003465777, + "loss": 1.9124, + "step": 1244 + }, + { + "epoch": 0.12158203125, + "grad_norm": 0.2503783106803894, + "learning_rate": 0.0004900387861048208, + "loss": 1.9426, + "step": 1245 + }, + { + "epoch": 0.1216796875, + "grad_norm": 0.36085277795791626, + "learning_rate": 0.0004900178503825942, + "loss": 1.9141, + "step": 1246 + }, + { + "epoch": 0.12177734375, + "grad_norm": 0.2831665873527527, + "learning_rate": 0.0004899968931819893, + "loss": 1.9515, + "step": 1247 + }, + { + "epoch": 0.121875, + "grad_norm": 0.26685988903045654, + "learning_rate": 0.0004899759145050996, + "loss": 1.9752, + "step": 1248 + }, + { + "epoch": 0.12197265625, + "grad_norm": 0.3139730393886566, + "learning_rate": 0.0004899549143540205, + "loss": 1.9298, + "step": 1249 + }, + { + "epoch": 0.1220703125, + "grad_norm": 0.20647141337394714, + "learning_rate": 0.0004899338927308497, + "loss": 1.9233, + "step": 1250 + }, + { + "epoch": 0.12216796875, + "grad_norm": 0.293544203042984, + "learning_rate": 0.0004899128496376871, + "loss": 1.9534, + "step": 1251 + }, + { + "epoch": 0.122265625, + "grad_norm": 0.28824836015701294, + "learning_rate": 0.0004898917850766349, + "loss": 1.96, + "step": 1252 + }, + { + "epoch": 0.12236328125, + "grad_norm": 0.23198992013931274, + "learning_rate": 0.0004898706990497971, + "loss": 1.9275, + "step": 1253 + }, + { + "epoch": 0.1224609375, + "grad_norm": 0.27970659732818604, + "learning_rate": 0.00048984959155928, + "loss": 1.8998, + "step": 1254 + }, + { + "epoch": 0.12255859375, + "grad_norm": 0.2355661541223526, + "learning_rate": 0.0004898284626071921, + "loss": 1.9313, + "step": 1255 + }, + { + "epoch": 0.12265625, + "grad_norm": 0.3456537127494812, + "learning_rate": 0.0004898073121956441, + "loss": 1.9164, + "step": 1256 + }, + { + "epoch": 0.12275390625, + "grad_norm": 0.32490289211273193, + "learning_rate": 0.0004897861403267486, + "loss": 1.9568, + "step": 1257 + }, + { + "epoch": 0.1228515625, + "grad_norm": 0.37740960717201233, + "learning_rate": 0.0004897649470026205, + "loss": 1.9618, + "step": 1258 + }, + { + "epoch": 0.12294921875, + "grad_norm": 0.34424281120300293, + "learning_rate": 0.0004897437322253767, + "loss": 1.9192, + "step": 1259 + }, + { + "epoch": 0.123046875, + "grad_norm": 0.1956419050693512, + "learning_rate": 0.0004897224959971368, + "loss": 1.9128, + "step": 1260 + }, + { + "epoch": 0.12314453125, + "grad_norm": 0.30533185601234436, + "learning_rate": 0.0004897012383200216, + "loss": 1.9378, + "step": 1261 + }, + { + "epoch": 0.1232421875, + "grad_norm": 0.28745076060295105, + "learning_rate": 0.0004896799591961548, + "loss": 1.9734, + "step": 1262 + }, + { + "epoch": 0.12333984375, + "grad_norm": 0.3063562512397766, + "learning_rate": 0.000489658658627662, + "loss": 1.9713, + "step": 1263 + }, + { + "epoch": 0.1234375, + "grad_norm": 0.26775872707366943, + "learning_rate": 0.000489637336616671, + "loss": 1.9305, + "step": 1264 + }, + { + "epoch": 0.12353515625, + "grad_norm": 0.2045804262161255, + "learning_rate": 0.0004896159931653114, + "loss": 1.9018, + "step": 1265 + }, + { + "epoch": 0.1236328125, + "grad_norm": 0.26752471923828125, + "learning_rate": 0.0004895946282757155, + "loss": 1.9334, + "step": 1266 + }, + { + "epoch": 0.12373046875, + "grad_norm": 0.3504626452922821, + "learning_rate": 0.0004895732419500174, + "loss": 1.9678, + "step": 1267 + }, + { + "epoch": 0.123828125, + "grad_norm": 0.2588043808937073, + "learning_rate": 0.0004895518341903534, + "loss": 1.9336, + "step": 1268 + }, + { + "epoch": 0.12392578125, + "grad_norm": 0.23393654823303223, + "learning_rate": 0.0004895304049988619, + "loss": 1.9386, + "step": 1269 + }, + { + "epoch": 0.1240234375, + "grad_norm": 0.3162434697151184, + "learning_rate": 0.0004895089543776834, + "loss": 1.9752, + "step": 1270 + }, + { + "epoch": 0.12412109375, + "grad_norm": 0.28987085819244385, + "learning_rate": 0.0004894874823289609, + "loss": 1.9346, + "step": 1271 + }, + { + "epoch": 0.12421875, + "grad_norm": 0.2837126553058624, + "learning_rate": 0.000489465988854839, + "loss": 1.9375, + "step": 1272 + }, + { + "epoch": 0.12431640625, + "grad_norm": 0.26402387022972107, + "learning_rate": 0.0004894444739574649, + "loss": 1.9235, + "step": 1273 + }, + { + "epoch": 0.1244140625, + "grad_norm": 0.21077674627304077, + "learning_rate": 0.0004894229376389875, + "loss": 1.9524, + "step": 1274 + }, + { + "epoch": 0.12451171875, + "grad_norm": 0.18328504264354706, + "learning_rate": 0.0004894013799015583, + "loss": 1.9496, + "step": 1275 + }, + { + "epoch": 0.124609375, + "grad_norm": 0.22258712351322174, + "learning_rate": 0.0004893798007473307, + "loss": 1.9661, + "step": 1276 + }, + { + "epoch": 0.12470703125, + "grad_norm": 0.20786426961421967, + "learning_rate": 0.0004893582001784604, + "loss": 1.9263, + "step": 1277 + }, + { + "epoch": 0.1248046875, + "grad_norm": 0.1959957331418991, + "learning_rate": 0.0004893365781971047, + "loss": 1.9525, + "step": 1278 + }, + { + "epoch": 0.12490234375, + "grad_norm": 0.2622356712818146, + "learning_rate": 0.0004893149348054238, + "loss": 1.9433, + "step": 1279 + }, + { + "epoch": 0.125, + "grad_norm": 0.2329346388578415, + "learning_rate": 0.0004892932700055797, + "loss": 1.9293, + "step": 1280 + }, + { + "epoch": 0.12509765625, + "grad_norm": 0.26422178745269775, + "learning_rate": 0.0004892715837997361, + "loss": 1.9252, + "step": 1281 + }, + { + "epoch": 0.1251953125, + "grad_norm": 0.30737435817718506, + "learning_rate": 0.0004892498761900597, + "loss": 1.9171, + "step": 1282 + }, + { + "epoch": 0.12529296875, + "grad_norm": 0.30651819705963135, + "learning_rate": 0.0004892281471787187, + "loss": 1.937, + "step": 1283 + }, + { + "epoch": 0.125390625, + "grad_norm": 0.23027795553207397, + "learning_rate": 0.0004892063967678838, + "loss": 1.9491, + "step": 1284 + }, + { + "epoch": 0.12548828125, + "grad_norm": 0.30681830644607544, + "learning_rate": 0.0004891846249597276, + "loss": 1.905, + "step": 1285 + }, + { + "epoch": 0.1255859375, + "grad_norm": 0.34682074189186096, + "learning_rate": 0.0004891628317564247, + "loss": 1.9157, + "step": 1286 + }, + { + "epoch": 0.12568359375, + "grad_norm": 0.26123619079589844, + "learning_rate": 0.0004891410171601523, + "loss": 1.9382, + "step": 1287 + }, + { + "epoch": 0.12578125, + "grad_norm": 0.32031431794166565, + "learning_rate": 0.0004891191811730894, + "loss": 1.9234, + "step": 1288 + }, + { + "epoch": 0.12587890625, + "grad_norm": 0.24917498230934143, + "learning_rate": 0.0004890973237974172, + "loss": 1.919, + "step": 1289 + }, + { + "epoch": 0.1259765625, + "grad_norm": 0.29161348938941956, + "learning_rate": 0.0004890754450353192, + "loss": 1.9405, + "step": 1290 + }, + { + "epoch": 0.12607421875, + "grad_norm": 0.31046807765960693, + "learning_rate": 0.0004890535448889807, + "loss": 1.9298, + "step": 1291 + }, + { + "epoch": 0.126171875, + "grad_norm": 0.30187463760375977, + "learning_rate": 0.0004890316233605893, + "loss": 1.9329, + "step": 1292 + }, + { + "epoch": 0.12626953125, + "grad_norm": 0.3013293445110321, + "learning_rate": 0.000489009680452335, + "loss": 1.9228, + "step": 1293 + }, + { + "epoch": 0.1263671875, + "grad_norm": 0.2492775321006775, + "learning_rate": 0.0004889877161664096, + "loss": 1.9565, + "step": 1294 + }, + { + "epoch": 0.12646484375, + "grad_norm": 0.29035332798957825, + "learning_rate": 0.000488965730505007, + "loss": 1.9184, + "step": 1295 + }, + { + "epoch": 0.1265625, + "grad_norm": 0.1983688622713089, + "learning_rate": 0.0004889437234703236, + "loss": 1.97, + "step": 1296 + }, + { + "epoch": 0.12666015625, + "grad_norm": 0.24074336886405945, + "learning_rate": 0.0004889216950645576, + "loss": 1.9289, + "step": 1297 + }, + { + "epoch": 0.1267578125, + "grad_norm": 0.2601425051689148, + "learning_rate": 0.0004888996452899093, + "loss": 1.9374, + "step": 1298 + }, + { + "epoch": 0.12685546875, + "grad_norm": 0.21484848856925964, + "learning_rate": 0.0004888775741485816, + "loss": 1.9075, + "step": 1299 + }, + { + "epoch": 0.126953125, + "grad_norm": 0.2370765060186386, + "learning_rate": 0.0004888554816427789, + "loss": 1.9131, + "step": 1300 + }, + { + "epoch": 0.12705078125, + "grad_norm": 0.22826816141605377, + "learning_rate": 0.0004888333677747082, + "loss": 1.9514, + "step": 1301 + }, + { + "epoch": 0.1271484375, + "grad_norm": 0.2141917198896408, + "learning_rate": 0.0004888112325465785, + "loss": 1.8406, + "step": 1302 + }, + { + "epoch": 0.12724609375, + "grad_norm": 0.22987738251686096, + "learning_rate": 0.0004887890759606008, + "loss": 1.9498, + "step": 1303 + }, + { + "epoch": 0.12734375, + "grad_norm": 0.2592550814151764, + "learning_rate": 0.0004887668980189885, + "loss": 1.9427, + "step": 1304 + }, + { + "epoch": 0.12744140625, + "grad_norm": 0.24788032472133636, + "learning_rate": 0.0004887446987239567, + "loss": 1.9718, + "step": 1305 + }, + { + "epoch": 0.1275390625, + "grad_norm": 0.3112640380859375, + "learning_rate": 0.0004887224780777233, + "loss": 1.972, + "step": 1306 + }, + { + "epoch": 0.12763671875, + "grad_norm": 0.2711445987224579, + "learning_rate": 0.0004887002360825077, + "loss": 1.9636, + "step": 1307 + }, + { + "epoch": 0.127734375, + "grad_norm": 0.19959181547164917, + "learning_rate": 0.0004886779727405318, + "loss": 1.9142, + "step": 1308 + }, + { + "epoch": 0.12783203125, + "grad_norm": 0.24054476618766785, + "learning_rate": 0.0004886556880540192, + "loss": 1.8724, + "step": 1309 + }, + { + "epoch": 0.1279296875, + "grad_norm": 0.21965253353118896, + "learning_rate": 0.0004886333820251963, + "loss": 1.9295, + "step": 1310 + }, + { + "epoch": 0.12802734375, + "grad_norm": 0.2383204847574234, + "learning_rate": 0.0004886110546562912, + "loss": 1.8945, + "step": 1311 + }, + { + "epoch": 0.128125, + "grad_norm": 0.20343053340911865, + "learning_rate": 0.0004885887059495342, + "loss": 1.9364, + "step": 1312 + }, + { + "epoch": 0.12822265625, + "grad_norm": 0.24133317172527313, + "learning_rate": 0.0004885663359071576, + "loss": 1.9023, + "step": 1313 + }, + { + "epoch": 0.1283203125, + "grad_norm": 0.27090492844581604, + "learning_rate": 0.0004885439445313962, + "loss": 1.9007, + "step": 1314 + }, + { + "epoch": 0.12841796875, + "grad_norm": 0.2153930515050888, + "learning_rate": 0.0004885215318244865, + "loss": 1.9475, + "step": 1315 + }, + { + "epoch": 0.128515625, + "grad_norm": 0.3661288022994995, + "learning_rate": 0.0004884990977886673, + "loss": 1.9858, + "step": 1316 + }, + { + "epoch": 0.12861328125, + "grad_norm": 0.4454726576805115, + "learning_rate": 0.0004884766424261798, + "loss": 1.9207, + "step": 1317 + }, + { + "epoch": 0.1287109375, + "grad_norm": 0.4249671995639801, + "learning_rate": 0.000488454165739267, + "loss": 1.9889, + "step": 1318 + }, + { + "epoch": 0.12880859375, + "grad_norm": 0.3520338237285614, + "learning_rate": 0.0004884316677301739, + "loss": 1.9564, + "step": 1319 + }, + { + "epoch": 0.12890625, + "grad_norm": 0.28052136301994324, + "learning_rate": 0.0004884091484011482, + "loss": 1.9677, + "step": 1320 + }, + { + "epoch": 0.12900390625, + "grad_norm": 0.31618887186050415, + "learning_rate": 0.0004883866077544391, + "loss": 1.9306, + "step": 1321 + }, + { + "epoch": 0.1291015625, + "grad_norm": 0.29120492935180664, + "learning_rate": 0.0004883640457922984, + "loss": 1.9205, + "step": 1322 + }, + { + "epoch": 0.12919921875, + "grad_norm": 0.23774267733097076, + "learning_rate": 0.0004883414625169797, + "loss": 1.9078, + "step": 1323 + }, + { + "epoch": 0.129296875, + "grad_norm": 0.3183814287185669, + "learning_rate": 0.0004883188579307389, + "loss": 1.7927, + "step": 1324 + }, + { + "epoch": 0.12939453125, + "grad_norm": 0.2845805585384369, + "learning_rate": 0.0004882962320358341, + "loss": 1.888, + "step": 1325 + }, + { + "epoch": 0.1294921875, + "grad_norm": 0.2877943515777588, + "learning_rate": 0.00048827358483452527, + "loss": 1.9307, + "step": 1326 + }, + { + "epoch": 0.12958984375, + "grad_norm": 0.2753991186618805, + "learning_rate": 0.0004882509163290748, + "loss": 1.9306, + "step": 1327 + }, + { + "epoch": 0.1296875, + "grad_norm": 0.2753439247608185, + "learning_rate": 0.00048822822652174705, + "loss": 1.9573, + "step": 1328 + }, + { + "epoch": 0.12978515625, + "grad_norm": 0.2690608501434326, + "learning_rate": 0.00048820551541480845, + "loss": 1.9229, + "step": 1329 + }, + { + "epoch": 0.1298828125, + "grad_norm": 0.2276460975408554, + "learning_rate": 0.00048818278301052776, + "loss": 1.9871, + "step": 1330 + }, + { + "epoch": 0.12998046875, + "grad_norm": 0.23562243580818176, + "learning_rate": 0.00048816002931117557, + "loss": 1.9489, + "step": 1331 + }, + { + "epoch": 0.130078125, + "grad_norm": 0.27421703934669495, + "learning_rate": 0.00048813725431902496, + "loss": 1.922, + "step": 1332 + }, + { + "epoch": 0.13017578125, + "grad_norm": 0.1969674974679947, + "learning_rate": 0.00048811445803635087, + "loss": 1.929, + "step": 1333 + }, + { + "epoch": 0.1302734375, + "grad_norm": 0.22562848031520844, + "learning_rate": 0.00048809164046543043, + "loss": 1.9183, + "step": 1334 + }, + { + "epoch": 0.13037109375, + "grad_norm": 0.2235979586839676, + "learning_rate": 0.0004880688016085429, + "loss": 1.9155, + "step": 1335 + }, + { + "epoch": 0.13046875, + "grad_norm": 0.23072493076324463, + "learning_rate": 0.00048804594146796977, + "loss": 1.9647, + "step": 1336 + }, + { + "epoch": 0.13056640625, + "grad_norm": 0.2528524398803711, + "learning_rate": 0.00048802306004599446, + "loss": 1.979, + "step": 1337 + }, + { + "epoch": 0.1306640625, + "grad_norm": 0.2586860954761505, + "learning_rate": 0.00048800015734490267, + "loss": 1.9371, + "step": 1338 + }, + { + "epoch": 0.13076171875, + "grad_norm": 0.21609428524971008, + "learning_rate": 0.0004879772333669822, + "loss": 1.9319, + "step": 1339 + }, + { + "epoch": 0.130859375, + "grad_norm": 0.28810006380081177, + "learning_rate": 0.0004879542881145229, + "loss": 1.9218, + "step": 1340 + }, + { + "epoch": 0.13095703125, + "grad_norm": 0.288849413394928, + "learning_rate": 0.0004879313215898168, + "loss": 1.929, + "step": 1341 + }, + { + "epoch": 0.1310546875, + "grad_norm": 0.3168569803237915, + "learning_rate": 0.00048790833379515813, + "loss": 1.9276, + "step": 1342 + }, + { + "epoch": 0.13115234375, + "grad_norm": 0.3000844120979309, + "learning_rate": 0.00048788532473284307, + "loss": 1.9337, + "step": 1343 + }, + { + "epoch": 0.13125, + "grad_norm": 0.2524905204772949, + "learning_rate": 0.00048786229440517004, + "loss": 1.9105, + "step": 1344 + }, + { + "epoch": 0.13134765625, + "grad_norm": 0.25065815448760986, + "learning_rate": 0.0004878392428144396, + "loss": 1.9161, + "step": 1345 + }, + { + "epoch": 0.1314453125, + "grad_norm": 0.3235273063182831, + "learning_rate": 0.00048781616996295443, + "loss": 1.9159, + "step": 1346 + }, + { + "epoch": 0.13154296875, + "grad_norm": 0.23726877570152283, + "learning_rate": 0.0004877930758530192, + "loss": 1.9063, + "step": 1347 + }, + { + "epoch": 0.131640625, + "grad_norm": 0.2991824150085449, + "learning_rate": 0.000487769960486941, + "loss": 1.9885, + "step": 1348 + }, + { + "epoch": 0.13173828125, + "grad_norm": 0.24819223582744598, + "learning_rate": 0.0004877468238670287, + "loss": 1.9311, + "step": 1349 + }, + { + "epoch": 0.1318359375, + "grad_norm": 0.23198595643043518, + "learning_rate": 0.00048772366599559333, + "loss": 1.9278, + "step": 1350 + }, + { + "epoch": 0.13193359375, + "grad_norm": 0.3012297451496124, + "learning_rate": 0.0004877004868749483, + "loss": 1.9258, + "step": 1351 + }, + { + "epoch": 0.13203125, + "grad_norm": 0.21619653701782227, + "learning_rate": 0.0004876772865074091, + "loss": 1.9121, + "step": 1352 + }, + { + "epoch": 0.13212890625, + "grad_norm": 0.24853789806365967, + "learning_rate": 0.00048765406489529315, + "loss": 1.9533, + "step": 1353 + }, + { + "epoch": 0.1322265625, + "grad_norm": 0.33258283138275146, + "learning_rate": 0.00048763082204091995, + "loss": 1.9424, + "step": 1354 + }, + { + "epoch": 0.13232421875, + "grad_norm": 0.3056190609931946, + "learning_rate": 0.00048760755794661137, + "loss": 1.9605, + "step": 1355 + }, + { + "epoch": 0.132421875, + "grad_norm": 0.20714829862117767, + "learning_rate": 0.00048758427261469135, + "loss": 1.9416, + "step": 1356 + }, + { + "epoch": 0.13251953125, + "grad_norm": 0.2573312520980835, + "learning_rate": 0.0004875609660474857, + "loss": 1.9198, + "step": 1357 + }, + { + "epoch": 0.1326171875, + "grad_norm": 0.24284306168556213, + "learning_rate": 0.0004875376382473228, + "loss": 1.9308, + "step": 1358 + }, + { + "epoch": 0.13271484375, + "grad_norm": 0.20341083407402039, + "learning_rate": 0.0004875142892165327, + "loss": 1.8933, + "step": 1359 + }, + { + "epoch": 0.1328125, + "grad_norm": 0.3155312240123749, + "learning_rate": 0.0004874909189574478, + "loss": 1.8825, + "step": 1360 + }, + { + "epoch": 0.13291015625, + "grad_norm": 0.2829397916793823, + "learning_rate": 0.0004874675274724026, + "loss": 1.9229, + "step": 1361 + }, + { + "epoch": 0.1330078125, + "grad_norm": 0.34132909774780273, + "learning_rate": 0.00048744411476373374, + "loss": 1.8935, + "step": 1362 + }, + { + "epoch": 0.13310546875, + "grad_norm": 0.41804420948028564, + "learning_rate": 0.0004874206808337799, + "loss": 1.9241, + "step": 1363 + }, + { + "epoch": 0.133203125, + "grad_norm": 0.3885767161846161, + "learning_rate": 0.0004873972256848819, + "loss": 1.9391, + "step": 1364 + }, + { + "epoch": 0.13330078125, + "grad_norm": 0.3422619104385376, + "learning_rate": 0.00048737374931938273, + "loss": 1.9113, + "step": 1365 + }, + { + "epoch": 0.1333984375, + "grad_norm": 0.33838555216789246, + "learning_rate": 0.00048735025173962746, + "loss": 1.8967, + "step": 1366 + }, + { + "epoch": 0.13349609375, + "grad_norm": 0.3656925857067108, + "learning_rate": 0.0004873267329479633, + "loss": 1.9405, + "step": 1367 + }, + { + "epoch": 0.13359375, + "grad_norm": 0.3086741268634796, + "learning_rate": 0.00048730319294673955, + "loss": 1.9518, + "step": 1368 + }, + { + "epoch": 0.13369140625, + "grad_norm": 0.24022363126277924, + "learning_rate": 0.0004872796317383077, + "loss": 1.9368, + "step": 1369 + }, + { + "epoch": 0.1337890625, + "grad_norm": 0.2861846089363098, + "learning_rate": 0.00048725604932502116, + "loss": 1.9186, + "step": 1370 + }, + { + "epoch": 0.13388671875, + "grad_norm": 0.29110145568847656, + "learning_rate": 0.0004872324457092358, + "loss": 1.8963, + "step": 1371 + }, + { + "epoch": 0.133984375, + "grad_norm": 0.2232580929994583, + "learning_rate": 0.0004872088208933093, + "loss": 1.929, + "step": 1372 + }, + { + "epoch": 0.13408203125, + "grad_norm": 0.3331020772457123, + "learning_rate": 0.0004871851748796016, + "loss": 1.936, + "step": 1373 + }, + { + "epoch": 0.1341796875, + "grad_norm": 0.2353833019733429, + "learning_rate": 0.00048716150767047473, + "loss": 1.9047, + "step": 1374 + }, + { + "epoch": 0.13427734375, + "grad_norm": 0.25953710079193115, + "learning_rate": 0.0004871378192682928, + "loss": 1.9812, + "step": 1375 + }, + { + "epoch": 0.134375, + "grad_norm": 0.24382802844047546, + "learning_rate": 0.0004871141096754222, + "loss": 1.9599, + "step": 1376 + }, + { + "epoch": 0.13447265625, + "grad_norm": 0.2496114820241928, + "learning_rate": 0.00048709037889423107, + "loss": 1.9094, + "step": 1377 + }, + { + "epoch": 0.1345703125, + "grad_norm": 0.26899224519729614, + "learning_rate": 0.0004870666269270901, + "loss": 1.9467, + "step": 1378 + }, + { + "epoch": 0.13466796875, + "grad_norm": 0.23490363359451294, + "learning_rate": 0.00048704285377637184, + "loss": 1.9682, + "step": 1379 + }, + { + "epoch": 0.134765625, + "grad_norm": 0.23190240561962128, + "learning_rate": 0.000487019059444451, + "loss": 1.9324, + "step": 1380 + }, + { + "epoch": 0.13486328125, + "grad_norm": 0.2239055186510086, + "learning_rate": 0.0004869952439337045, + "loss": 1.995, + "step": 1381 + }, + { + "epoch": 0.1349609375, + "grad_norm": 0.23679253458976746, + "learning_rate": 0.00048697140724651114, + "loss": 1.8685, + "step": 1382 + }, + { + "epoch": 0.13505859375, + "grad_norm": 0.2584695816040039, + "learning_rate": 0.0004869475493852521, + "loss": 1.9421, + "step": 1383 + }, + { + "epoch": 0.13515625, + "grad_norm": 0.219711571931839, + "learning_rate": 0.0004869236703523106, + "loss": 1.9326, + "step": 1384 + }, + { + "epoch": 0.13525390625, + "grad_norm": 0.2176481932401657, + "learning_rate": 0.00048689977015007197, + "loss": 1.8959, + "step": 1385 + }, + { + "epoch": 0.1353515625, + "grad_norm": 0.22419661283493042, + "learning_rate": 0.0004868758487809235, + "loss": 1.9095, + "step": 1386 + }, + { + "epoch": 0.13544921875, + "grad_norm": 0.255984365940094, + "learning_rate": 0.0004868519062472548, + "loss": 1.9756, + "step": 1387 + }, + { + "epoch": 0.135546875, + "grad_norm": 0.217428520321846, + "learning_rate": 0.0004868279425514575, + "loss": 1.9481, + "step": 1388 + }, + { + "epoch": 0.13564453125, + "grad_norm": 0.19990497827529907, + "learning_rate": 0.00048680395769592534, + "loss": 1.9279, + "step": 1389 + }, + { + "epoch": 0.1357421875, + "grad_norm": 0.2209380716085434, + "learning_rate": 0.00048677995168305426, + "loss": 1.9148, + "step": 1390 + }, + { + "epoch": 0.13583984375, + "grad_norm": 0.25395241379737854, + "learning_rate": 0.00048675592451524226, + "loss": 1.9688, + "step": 1391 + }, + { + "epoch": 0.1359375, + "grad_norm": 0.1944066882133484, + "learning_rate": 0.0004867318761948892, + "loss": 1.9005, + "step": 1392 + }, + { + "epoch": 0.13603515625, + "grad_norm": 0.22825607657432556, + "learning_rate": 0.0004867078067243977, + "loss": 1.9538, + "step": 1393 + }, + { + "epoch": 0.1361328125, + "grad_norm": 0.23758959770202637, + "learning_rate": 0.0004866837161061717, + "loss": 1.9887, + "step": 1394 + }, + { + "epoch": 0.13623046875, + "grad_norm": 0.26341712474823, + "learning_rate": 0.00048665960434261796, + "loss": 1.9217, + "step": 1395 + }, + { + "epoch": 0.136328125, + "grad_norm": 0.283316045999527, + "learning_rate": 0.0004866354714361448, + "loss": 1.9576, + "step": 1396 + }, + { + "epoch": 0.13642578125, + "grad_norm": 0.32178181409835815, + "learning_rate": 0.000486611317389163, + "loss": 1.9651, + "step": 1397 + }, + { + "epoch": 0.1365234375, + "grad_norm": 0.4065782129764557, + "learning_rate": 0.0004865871422040853, + "loss": 1.897, + "step": 1398 + }, + { + "epoch": 0.13662109375, + "grad_norm": 0.3661766052246094, + "learning_rate": 0.00048656294588332653, + "loss": 1.9472, + "step": 1399 + }, + { + "epoch": 0.13671875, + "grad_norm": 0.24434155225753784, + "learning_rate": 0.0004865387284293038, + "loss": 1.9179, + "step": 1400 + }, + { + "epoch": 0.13681640625, + "grad_norm": 0.4203675091266632, + "learning_rate": 0.0004865144898444361, + "loss": 1.9583, + "step": 1401 + }, + { + "epoch": 0.1369140625, + "grad_norm": 0.4013305902481079, + "learning_rate": 0.0004864902301311448, + "loss": 1.9322, + "step": 1402 + }, + { + "epoch": 0.13701171875, + "grad_norm": 0.30299681425094604, + "learning_rate": 0.00048646594929185313, + "loss": 1.9152, + "step": 1403 + }, + { + "epoch": 0.137109375, + "grad_norm": 0.30768755078315735, + "learning_rate": 0.0004864416473289865, + "loss": 1.966, + "step": 1404 + }, + { + "epoch": 0.13720703125, + "grad_norm": 0.2759581506252289, + "learning_rate": 0.00048641732424497256, + "loss": 1.9211, + "step": 1405 + }, + { + "epoch": 0.1373046875, + "grad_norm": 0.2976212799549103, + "learning_rate": 0.00048639298004224087, + "loss": 1.9376, + "step": 1406 + }, + { + "epoch": 0.13740234375, + "grad_norm": 0.21789424121379852, + "learning_rate": 0.0004863686147232232, + "loss": 1.994, + "step": 1407 + }, + { + "epoch": 0.1375, + "grad_norm": 0.2543160617351532, + "learning_rate": 0.0004863442282903535, + "loss": 1.9153, + "step": 1408 + }, + { + "epoch": 0.13759765625, + "grad_norm": 0.2789456248283386, + "learning_rate": 0.0004863198207460678, + "loss": 1.9238, + "step": 1409 + }, + { + "epoch": 0.1376953125, + "grad_norm": 0.21548853814601898, + "learning_rate": 0.0004862953920928041, + "loss": 1.9003, + "step": 1410 + }, + { + "epoch": 0.13779296875, + "grad_norm": 0.2607603967189789, + "learning_rate": 0.00048627094233300266, + "loss": 1.951, + "step": 1411 + }, + { + "epoch": 0.137890625, + "grad_norm": 0.21658223867416382, + "learning_rate": 0.00048624647146910575, + "loss": 1.9818, + "step": 1412 + }, + { + "epoch": 0.13798828125, + "grad_norm": 0.24333196878433228, + "learning_rate": 0.00048622197950355786, + "loss": 1.8933, + "step": 1413 + }, + { + "epoch": 0.1380859375, + "grad_norm": 0.22348099946975708, + "learning_rate": 0.0004861974664388054, + "loss": 1.9302, + "step": 1414 + }, + { + "epoch": 0.13818359375, + "grad_norm": 0.1976805180311203, + "learning_rate": 0.0004861729322772971, + "loss": 1.9044, + "step": 1415 + }, + { + "epoch": 0.13828125, + "grad_norm": 0.22617708146572113, + "learning_rate": 0.00048614837702148374, + "loss": 1.9445, + "step": 1416 + }, + { + "epoch": 0.13837890625, + "grad_norm": 0.2243851125240326, + "learning_rate": 0.0004861238006738181, + "loss": 1.9215, + "step": 1417 + }, + { + "epoch": 0.1384765625, + "grad_norm": 0.21161694824695587, + "learning_rate": 0.0004860992032367551, + "loss": 1.9443, + "step": 1418 + }, + { + "epoch": 0.13857421875, + "grad_norm": 0.22872799634933472, + "learning_rate": 0.0004860745847127519, + "loss": 1.8174, + "step": 1419 + }, + { + "epoch": 0.138671875, + "grad_norm": 0.27401795983314514, + "learning_rate": 0.0004860499451042677, + "loss": 1.919, + "step": 1420 + }, + { + "epoch": 0.13876953125, + "grad_norm": 0.2309410721063614, + "learning_rate": 0.00048602528441376365, + "loss": 1.9025, + "step": 1421 + }, + { + "epoch": 0.1388671875, + "grad_norm": 0.21750116348266602, + "learning_rate": 0.0004860006026437033, + "loss": 1.9242, + "step": 1422 + }, + { + "epoch": 0.13896484375, + "grad_norm": 0.24858328700065613, + "learning_rate": 0.0004859758997965519, + "loss": 1.9562, + "step": 1423 + }, + { + "epoch": 0.1390625, + "grad_norm": 0.23306035995483398, + "learning_rate": 0.00048595117587477724, + "loss": 1.9504, + "step": 1424 + }, + { + "epoch": 0.13916015625, + "grad_norm": 0.24605897068977356, + "learning_rate": 0.00048592643088084904, + "loss": 1.9502, + "step": 1425 + }, + { + "epoch": 0.1392578125, + "grad_norm": 0.23687416315078735, + "learning_rate": 0.0004859016648172389, + "loss": 1.9035, + "step": 1426 + }, + { + "epoch": 0.13935546875, + "grad_norm": 0.33239808678627014, + "learning_rate": 0.00048587687768642104, + "loss": 1.9469, + "step": 1427 + }, + { + "epoch": 0.139453125, + "grad_norm": 0.3443644344806671, + "learning_rate": 0.00048585206949087117, + "loss": 1.9126, + "step": 1428 + }, + { + "epoch": 0.13955078125, + "grad_norm": 0.4044901430606842, + "learning_rate": 0.0004858272402330676, + "loss": 1.928, + "step": 1429 + }, + { + "epoch": 0.1396484375, + "grad_norm": 0.3553960919380188, + "learning_rate": 0.0004858023899154905, + "loss": 1.9782, + "step": 1430 + }, + { + "epoch": 0.13974609375, + "grad_norm": 0.24459035694599152, + "learning_rate": 0.00048577751854062207, + "loss": 1.9228, + "step": 1431 + }, + { + "epoch": 0.13984375, + "grad_norm": 0.26586851477622986, + "learning_rate": 0.000485752626110947, + "loss": 1.9607, + "step": 1432 + }, + { + "epoch": 0.13994140625, + "grad_norm": 0.2949119210243225, + "learning_rate": 0.0004857277126289516, + "loss": 1.9418, + "step": 1433 + }, + { + "epoch": 0.1400390625, + "grad_norm": 0.23309734463691711, + "learning_rate": 0.0004857027780971246, + "loss": 1.9036, + "step": 1434 + }, + { + "epoch": 0.14013671875, + "grad_norm": 0.27574148774147034, + "learning_rate": 0.00048567782251795674, + "loss": 1.883, + "step": 1435 + }, + { + "epoch": 0.140234375, + "grad_norm": 0.2791823446750641, + "learning_rate": 0.0004856528458939409, + "loss": 1.9021, + "step": 1436 + }, + { + "epoch": 0.14033203125, + "grad_norm": 0.3588283061981201, + "learning_rate": 0.00048562784822757187, + "loss": 1.9417, + "step": 1437 + }, + { + "epoch": 0.1404296875, + "grad_norm": 0.3069411814212799, + "learning_rate": 0.00048560282952134687, + "loss": 1.9055, + "step": 1438 + }, + { + "epoch": 0.14052734375, + "grad_norm": 0.3290352523326874, + "learning_rate": 0.00048557778977776495, + "loss": 1.9674, + "step": 1439 + }, + { + "epoch": 0.140625, + "grad_norm": 0.21290095150470734, + "learning_rate": 0.0004855527289993274, + "loss": 1.9099, + "step": 1440 + }, + { + "epoch": 0.14072265625, + "grad_norm": 0.28244757652282715, + "learning_rate": 0.00048552764718853755, + "loss": 1.9821, + "step": 1441 + }, + { + "epoch": 0.1408203125, + "grad_norm": 0.2707923948764801, + "learning_rate": 0.0004855025443479008, + "loss": 1.92, + "step": 1442 + }, + { + "epoch": 0.14091796875, + "grad_norm": 0.2617279589176178, + "learning_rate": 0.0004854774204799248, + "loss": 1.9425, + "step": 1443 + }, + { + "epoch": 0.141015625, + "grad_norm": 0.2797471880912781, + "learning_rate": 0.00048545227558711914, + "loss": 1.9239, + "step": 1444 + }, + { + "epoch": 0.14111328125, + "grad_norm": 0.21120323240756989, + "learning_rate": 0.00048542710967199563, + "loss": 1.9323, + "step": 1445 + }, + { + "epoch": 0.1412109375, + "grad_norm": 0.23532937467098236, + "learning_rate": 0.00048540192273706807, + "loss": 1.9033, + "step": 1446 + }, + { + "epoch": 0.14130859375, + "grad_norm": 0.26372474431991577, + "learning_rate": 0.00048537671478485233, + "loss": 1.9109, + "step": 1447 + }, + { + "epoch": 0.14140625, + "grad_norm": 0.2063819169998169, + "learning_rate": 0.0004853514858178667, + "loss": 1.8868, + "step": 1448 + }, + { + "epoch": 0.14150390625, + "grad_norm": 0.2495461255311966, + "learning_rate": 0.0004853262358386311, + "loss": 1.9261, + "step": 1449 + }, + { + "epoch": 0.1416015625, + "grad_norm": 0.24477644264698029, + "learning_rate": 0.00048530096484966786, + "loss": 1.8921, + "step": 1450 + }, + { + "epoch": 0.14169921875, + "grad_norm": 0.20758818089962006, + "learning_rate": 0.0004852756728535013, + "loss": 1.9274, + "step": 1451 + }, + { + "epoch": 0.141796875, + "grad_norm": 0.24920731782913208, + "learning_rate": 0.000485250359852658, + "loss": 1.9203, + "step": 1452 + }, + { + "epoch": 0.14189453125, + "grad_norm": 0.27062875032424927, + "learning_rate": 0.00048522502584966626, + "loss": 1.9225, + "step": 1453 + }, + { + "epoch": 0.1419921875, + "grad_norm": 0.26489493250846863, + "learning_rate": 0.00048519967084705696, + "loss": 1.9797, + "step": 1454 + }, + { + "epoch": 0.14208984375, + "grad_norm": 0.2739894688129425, + "learning_rate": 0.0004851742948473627, + "loss": 1.8824, + "step": 1455 + }, + { + "epoch": 0.1421875, + "grad_norm": 0.24199414253234863, + "learning_rate": 0.0004851488978531184, + "loss": 1.9439, + "step": 1456 + }, + { + "epoch": 0.14228515625, + "grad_norm": 0.21669942140579224, + "learning_rate": 0.00048512347986686094, + "loss": 1.8901, + "step": 1457 + }, + { + "epoch": 0.1423828125, + "grad_norm": 0.27475011348724365, + "learning_rate": 0.0004850980408911294, + "loss": 1.877, + "step": 1458 + }, + { + "epoch": 0.14248046875, + "grad_norm": 0.24100442230701447, + "learning_rate": 0.0004850725809284648, + "loss": 1.9286, + "step": 1459 + }, + { + "epoch": 0.142578125, + "grad_norm": 0.24343852698802948, + "learning_rate": 0.0004850470999814105, + "loss": 1.9503, + "step": 1460 + }, + { + "epoch": 0.14267578125, + "grad_norm": 0.22666993737220764, + "learning_rate": 0.0004850215980525118, + "loss": 1.8722, + "step": 1461 + }, + { + "epoch": 0.1427734375, + "grad_norm": 0.22877848148345947, + "learning_rate": 0.00048499607514431597, + "loss": 1.8726, + "step": 1462 + }, + { + "epoch": 0.14287109375, + "grad_norm": 0.21896860003471375, + "learning_rate": 0.0004849705312593727, + "loss": 1.9109, + "step": 1463 + }, + { + "epoch": 0.14296875, + "grad_norm": 0.2896305024623871, + "learning_rate": 0.00048494496640023356, + "loss": 1.9293, + "step": 1464 + }, + { + "epoch": 0.14306640625, + "grad_norm": 0.2690875232219696, + "learning_rate": 0.00048491938056945214, + "loss": 1.9191, + "step": 1465 + }, + { + "epoch": 0.1431640625, + "grad_norm": 0.27440106868743896, + "learning_rate": 0.0004848937737695844, + "loss": 1.9729, + "step": 1466 + }, + { + "epoch": 0.14326171875, + "grad_norm": 0.26410380005836487, + "learning_rate": 0.000484868146003188, + "loss": 1.8782, + "step": 1467 + }, + { + "epoch": 0.143359375, + "grad_norm": 0.2948254644870758, + "learning_rate": 0.00048484249727282326, + "loss": 1.9726, + "step": 1468 + }, + { + "epoch": 0.14345703125, + "grad_norm": 0.34213653206825256, + "learning_rate": 0.000484816827581052, + "loss": 1.8893, + "step": 1469 + }, + { + "epoch": 0.1435546875, + "grad_norm": 0.2864801287651062, + "learning_rate": 0.0004847911369304385, + "loss": 1.9717, + "step": 1470 + }, + { + "epoch": 0.14365234375, + "grad_norm": 0.23252104222774506, + "learning_rate": 0.00048476542532354896, + "loss": 1.9256, + "step": 1471 + }, + { + "epoch": 0.14375, + "grad_norm": 0.24983392655849457, + "learning_rate": 0.00048473969276295175, + "loss": 1.9275, + "step": 1472 + }, + { + "epoch": 0.14384765625, + "grad_norm": 0.24153786897659302, + "learning_rate": 0.0004847139392512175, + "loss": 1.8821, + "step": 1473 + }, + { + "epoch": 0.1439453125, + "grad_norm": 0.2148951143026352, + "learning_rate": 0.00048468816479091853, + "loss": 1.9125, + "step": 1474 + }, + { + "epoch": 0.14404296875, + "grad_norm": 0.20793981850147247, + "learning_rate": 0.00048466236938462956, + "loss": 1.9336, + "step": 1475 + }, + { + "epoch": 0.144140625, + "grad_norm": 0.22007182240486145, + "learning_rate": 0.0004846365530349273, + "loss": 1.9383, + "step": 1476 + }, + { + "epoch": 0.14423828125, + "grad_norm": 0.21164564788341522, + "learning_rate": 0.00048461071574439063, + "loss": 1.9188, + "step": 1477 + }, + { + "epoch": 0.1443359375, + "grad_norm": 0.26241186261177063, + "learning_rate": 0.00048458485751560045, + "loss": 1.9348, + "step": 1478 + }, + { + "epoch": 0.14443359375, + "grad_norm": 0.19936083257198334, + "learning_rate": 0.00048455897835113974, + "loss": 1.8605, + "step": 1479 + }, + { + "epoch": 0.14453125, + "grad_norm": 0.2293226420879364, + "learning_rate": 0.0004845330782535937, + "loss": 1.8666, + "step": 1480 + }, + { + "epoch": 0.14462890625, + "grad_norm": 0.27344003319740295, + "learning_rate": 0.0004845071572255494, + "loss": 1.9281, + "step": 1481 + }, + { + "epoch": 0.1447265625, + "grad_norm": 0.2886092960834503, + "learning_rate": 0.0004844812152695961, + "loss": 1.9106, + "step": 1482 + }, + { + "epoch": 0.14482421875, + "grad_norm": 0.345197856426239, + "learning_rate": 0.00048445525238832526, + "loss": 1.8825, + "step": 1483 + }, + { + "epoch": 0.144921875, + "grad_norm": 0.32034623622894287, + "learning_rate": 0.0004844292685843303, + "loss": 1.9103, + "step": 1484 + }, + { + "epoch": 0.14501953125, + "grad_norm": 0.30518102645874023, + "learning_rate": 0.00048440326386020685, + "loss": 1.9614, + "step": 1485 + }, + { + "epoch": 0.1451171875, + "grad_norm": 0.28948330879211426, + "learning_rate": 0.00048437723821855244, + "loss": 1.8966, + "step": 1486 + }, + { + "epoch": 0.14521484375, + "grad_norm": 0.28401121497154236, + "learning_rate": 0.00048435119166196684, + "loss": 1.9455, + "step": 1487 + }, + { + "epoch": 0.1453125, + "grad_norm": 0.2352132648229599, + "learning_rate": 0.0004843251241930519, + "loss": 1.9285, + "step": 1488 + }, + { + "epoch": 0.14541015625, + "grad_norm": 0.231564462184906, + "learning_rate": 0.0004842990358144115, + "loss": 1.9204, + "step": 1489 + }, + { + "epoch": 0.1455078125, + "grad_norm": 0.2393004596233368, + "learning_rate": 0.00048427292652865165, + "loss": 1.9449, + "step": 1490 + }, + { + "epoch": 0.14560546875, + "grad_norm": 0.2513711154460907, + "learning_rate": 0.00048424679633838046, + "loss": 1.9092, + "step": 1491 + }, + { + "epoch": 0.145703125, + "grad_norm": 0.22026249766349792, + "learning_rate": 0.000484220645246208, + "loss": 1.9439, + "step": 1492 + }, + { + "epoch": 0.14580078125, + "grad_norm": 0.29993200302124023, + "learning_rate": 0.0004841944732547467, + "loss": 1.8934, + "step": 1493 + }, + { + "epoch": 0.1458984375, + "grad_norm": 0.27871406078338623, + "learning_rate": 0.0004841682803666108, + "loss": 1.8947, + "step": 1494 + }, + { + "epoch": 0.14599609375, + "grad_norm": 0.21927416324615479, + "learning_rate": 0.0004841420665844166, + "loss": 1.9472, + "step": 1495 + }, + { + "epoch": 0.14609375, + "grad_norm": 0.24884743988513947, + "learning_rate": 0.0004841158319107829, + "loss": 1.9099, + "step": 1496 + }, + { + "epoch": 0.14619140625, + "grad_norm": 0.256428986787796, + "learning_rate": 0.0004840895763483302, + "loss": 1.9266, + "step": 1497 + }, + { + "epoch": 0.1462890625, + "grad_norm": 0.2777554392814636, + "learning_rate": 0.0004840632998996812, + "loss": 1.9764, + "step": 1498 + }, + { + "epoch": 0.14638671875, + "grad_norm": 0.3199704587459564, + "learning_rate": 0.00048403700256746066, + "loss": 1.9391, + "step": 1499 + }, + { + "epoch": 0.146484375, + "grad_norm": 0.33837810158729553, + "learning_rate": 0.00048401068435429545, + "loss": 1.9131, + "step": 1500 + }, + { + "epoch": 0.14658203125, + "grad_norm": 0.25376418232917786, + "learning_rate": 0.00048398434526281456, + "loss": 1.9391, + "step": 1501 + }, + { + "epoch": 0.1466796875, + "grad_norm": 0.2878507673740387, + "learning_rate": 0.00048395798529564897, + "loss": 1.9226, + "step": 1502 + }, + { + "epoch": 0.14677734375, + "grad_norm": 0.3166121244430542, + "learning_rate": 0.0004839316044554319, + "loss": 1.9024, + "step": 1503 + }, + { + "epoch": 0.146875, + "grad_norm": 0.30799803137779236, + "learning_rate": 0.00048390520274479857, + "loss": 1.9076, + "step": 1504 + }, + { + "epoch": 0.14697265625, + "grad_norm": 0.3012866675853729, + "learning_rate": 0.0004838787801663861, + "loss": 1.9282, + "step": 1505 + }, + { + "epoch": 0.1470703125, + "grad_norm": 0.29956120252609253, + "learning_rate": 0.000483852336722834, + "loss": 1.8934, + "step": 1506 + }, + { + "epoch": 0.14716796875, + "grad_norm": 0.2174024134874344, + "learning_rate": 0.0004838258724167838, + "loss": 1.9434, + "step": 1507 + }, + { + "epoch": 0.147265625, + "grad_norm": 0.31407755613327026, + "learning_rate": 0.0004837993872508789, + "loss": 1.8654, + "step": 1508 + }, + { + "epoch": 0.14736328125, + "grad_norm": 0.22638960182666779, + "learning_rate": 0.00048377288122776506, + "loss": 1.937, + "step": 1509 + }, + { + "epoch": 0.1474609375, + "grad_norm": 0.26045146584510803, + "learning_rate": 0.0004837463543500899, + "loss": 1.8986, + "step": 1510 + }, + { + "epoch": 0.14755859375, + "grad_norm": 0.2935525178909302, + "learning_rate": 0.0004837198066205034, + "loss": 1.9554, + "step": 1511 + }, + { + "epoch": 0.14765625, + "grad_norm": 0.20349502563476562, + "learning_rate": 0.0004836932380416571, + "loss": 1.893, + "step": 1512 + }, + { + "epoch": 0.14775390625, + "grad_norm": 0.28056201338768005, + "learning_rate": 0.0004836666486162053, + "loss": 1.983, + "step": 1513 + }, + { + "epoch": 0.1478515625, + "grad_norm": 0.27571672201156616, + "learning_rate": 0.00048364003834680395, + "loss": 1.9399, + "step": 1514 + }, + { + "epoch": 0.14794921875, + "grad_norm": 0.28008097410202026, + "learning_rate": 0.0004836134072361111, + "loss": 1.9549, + "step": 1515 + }, + { + "epoch": 0.148046875, + "grad_norm": 0.2211456298828125, + "learning_rate": 0.00048358675528678704, + "loss": 1.9245, + "step": 1516 + }, + { + "epoch": 0.14814453125, + "grad_norm": 0.259030818939209, + "learning_rate": 0.00048356008250149404, + "loss": 1.8919, + "step": 1517 + }, + { + "epoch": 0.1482421875, + "grad_norm": 0.2772839665412903, + "learning_rate": 0.00048353338888289645, + "loss": 1.9386, + "step": 1518 + }, + { + "epoch": 0.14833984375, + "grad_norm": 0.3579663336277008, + "learning_rate": 0.00048350667443366075, + "loss": 1.9375, + "step": 1519 + }, + { + "epoch": 0.1484375, + "grad_norm": 0.37625831365585327, + "learning_rate": 0.0004834799391564555, + "loss": 1.9508, + "step": 1520 + }, + { + "epoch": 0.14853515625, + "grad_norm": 0.23583106696605682, + "learning_rate": 0.0004834531830539512, + "loss": 1.8934, + "step": 1521 + }, + { + "epoch": 0.1486328125, + "grad_norm": 0.2464006394147873, + "learning_rate": 0.00048342640612882075, + "loss": 1.8932, + "step": 1522 + }, + { + "epoch": 0.14873046875, + "grad_norm": 0.30233830213546753, + "learning_rate": 0.0004833996083837388, + "loss": 1.9117, + "step": 1523 + }, + { + "epoch": 0.148828125, + "grad_norm": 0.23098327219486237, + "learning_rate": 0.0004833727898213822, + "loss": 1.9229, + "step": 1524 + }, + { + "epoch": 0.14892578125, + "grad_norm": 0.2418309450149536, + "learning_rate": 0.00048334595044442996, + "loss": 1.9369, + "step": 1525 + }, + { + "epoch": 0.1490234375, + "grad_norm": 0.2360529899597168, + "learning_rate": 0.000483319090255563, + "loss": 1.8828, + "step": 1526 + }, + { + "epoch": 0.14912109375, + "grad_norm": 0.24173954129219055, + "learning_rate": 0.0004832922092574645, + "loss": 1.9349, + "step": 1527 + }, + { + "epoch": 0.14921875, + "grad_norm": 0.21543891727924347, + "learning_rate": 0.0004832653074528196, + "loss": 1.9317, + "step": 1528 + }, + { + "epoch": 0.14931640625, + "grad_norm": 0.21203261613845825, + "learning_rate": 0.0004832383848443155, + "loss": 1.9261, + "step": 1529 + }, + { + "epoch": 0.1494140625, + "grad_norm": 0.19579768180847168, + "learning_rate": 0.0004832114414346417, + "loss": 1.878, + "step": 1530 + }, + { + "epoch": 0.14951171875, + "grad_norm": 0.23105336725711823, + "learning_rate": 0.0004831844772264894, + "loss": 1.8957, + "step": 1531 + }, + { + "epoch": 0.149609375, + "grad_norm": 0.1797371655702591, + "learning_rate": 0.0004831574922225521, + "loss": 1.9036, + "step": 1532 + }, + { + "epoch": 0.14970703125, + "grad_norm": 0.22427886724472046, + "learning_rate": 0.00048313048642552553, + "loss": 1.9143, + "step": 1533 + }, + { + "epoch": 0.1498046875, + "grad_norm": 0.2837050259113312, + "learning_rate": 0.00048310345983810723, + "loss": 1.9411, + "step": 1534 + }, + { + "epoch": 0.14990234375, + "grad_norm": 0.25001662969589233, + "learning_rate": 0.0004830764124629969, + "loss": 1.9157, + "step": 1535 + }, + { + "epoch": 0.15, + "grad_norm": 0.21143823862075806, + "learning_rate": 0.00048304934430289635, + "loss": 1.9654, + "step": 1536 + }, + { + "epoch": 0.15009765625, + "grad_norm": 0.27027738094329834, + "learning_rate": 0.00048302225536050954, + "loss": 1.8961, + "step": 1537 + }, + { + "epoch": 0.1501953125, + "grad_norm": 0.2757401168346405, + "learning_rate": 0.00048299514563854225, + "loss": 1.9537, + "step": 1538 + }, + { + "epoch": 0.15029296875, + "grad_norm": 0.30425819754600525, + "learning_rate": 0.0004829680151397025, + "loss": 1.8713, + "step": 1539 + }, + { + "epoch": 0.150390625, + "grad_norm": 0.3185950815677643, + "learning_rate": 0.0004829408638667006, + "loss": 1.987, + "step": 1540 + }, + { + "epoch": 0.15048828125, + "grad_norm": 0.29616236686706543, + "learning_rate": 0.0004829136918222486, + "loss": 1.9329, + "step": 1541 + }, + { + "epoch": 0.1505859375, + "grad_norm": 0.2886502146720886, + "learning_rate": 0.00048288649900906066, + "loss": 1.9178, + "step": 1542 + }, + { + "epoch": 0.15068359375, + "grad_norm": 0.27268344163894653, + "learning_rate": 0.0004828592854298532, + "loss": 1.941, + "step": 1543 + }, + { + "epoch": 0.15078125, + "grad_norm": 0.27281153202056885, + "learning_rate": 0.00048283205108734463, + "loss": 1.9668, + "step": 1544 + }, + { + "epoch": 0.15087890625, + "grad_norm": 0.34445685148239136, + "learning_rate": 0.0004828047959842554, + "loss": 1.9153, + "step": 1545 + }, + { + "epoch": 0.1509765625, + "grad_norm": 0.3492426872253418, + "learning_rate": 0.0004827775201233079, + "loss": 1.9123, + "step": 1546 + }, + { + "epoch": 0.15107421875, + "grad_norm": 0.23972700536251068, + "learning_rate": 0.000482750223507227, + "loss": 1.9189, + "step": 1547 + }, + { + "epoch": 0.151171875, + "grad_norm": 0.28324854373931885, + "learning_rate": 0.0004827229061387393, + "loss": 1.9113, + "step": 1548 + }, + { + "epoch": 0.15126953125, + "grad_norm": 0.26321306824684143, + "learning_rate": 0.0004826955680205734, + "loss": 1.9568, + "step": 1549 + }, + { + "epoch": 0.1513671875, + "grad_norm": 0.26853737235069275, + "learning_rate": 0.0004826682091554604, + "loss": 1.8798, + "step": 1550 + }, + { + "epoch": 0.15146484375, + "grad_norm": 0.32173535227775574, + "learning_rate": 0.00048264082954613296, + "loss": 1.9564, + "step": 1551 + }, + { + "epoch": 0.1515625, + "grad_norm": 0.2380555272102356, + "learning_rate": 0.0004826134291953263, + "loss": 1.883, + "step": 1552 + }, + { + "epoch": 0.15166015625, + "grad_norm": 0.251919150352478, + "learning_rate": 0.0004825860081057773, + "loss": 1.8876, + "step": 1553 + }, + { + "epoch": 0.1517578125, + "grad_norm": 0.26018527150154114, + "learning_rate": 0.00048255856628022514, + "loss": 1.9115, + "step": 1554 + }, + { + "epoch": 0.15185546875, + "grad_norm": 0.21712715923786163, + "learning_rate": 0.00048253110372141104, + "loss": 1.9097, + "step": 1555 + }, + { + "epoch": 0.151953125, + "grad_norm": 0.2382095605134964, + "learning_rate": 0.0004825036204320782, + "loss": 1.937, + "step": 1556 + }, + { + "epoch": 0.15205078125, + "grad_norm": 0.231217160820961, + "learning_rate": 0.000482476116414972, + "loss": 1.9344, + "step": 1557 + }, + { + "epoch": 0.1521484375, + "grad_norm": 0.20458881556987762, + "learning_rate": 0.0004824485916728398, + "loss": 1.9036, + "step": 1558 + }, + { + "epoch": 0.15224609375, + "grad_norm": 0.205721914768219, + "learning_rate": 0.00048242104620843125, + "loss": 1.8998, + "step": 1559 + }, + { + "epoch": 0.15234375, + "grad_norm": 0.23993369936943054, + "learning_rate": 0.0004823934800244976, + "loss": 1.8706, + "step": 1560 + }, + { + "epoch": 0.15244140625, + "grad_norm": 0.23396044969558716, + "learning_rate": 0.00048236589312379276, + "loss": 1.8834, + "step": 1561 + }, + { + "epoch": 0.1525390625, + "grad_norm": 0.24282585084438324, + "learning_rate": 0.0004823382855090723, + "loss": 1.8873, + "step": 1562 + }, + { + "epoch": 0.15263671875, + "grad_norm": 0.25903937220573425, + "learning_rate": 0.00048231065718309387, + "loss": 1.9357, + "step": 1563 + }, + { + "epoch": 0.152734375, + "grad_norm": 0.24471600353717804, + "learning_rate": 0.0004822830081486175, + "loss": 1.8975, + "step": 1564 + }, + { + "epoch": 0.15283203125, + "grad_norm": 0.21125410497188568, + "learning_rate": 0.00048225533840840493, + "loss": 1.9513, + "step": 1565 + }, + { + "epoch": 0.1529296875, + "grad_norm": 0.22737543284893036, + "learning_rate": 0.00048222764796522017, + "loss": 1.8962, + "step": 1566 + }, + { + "epoch": 0.15302734375, + "grad_norm": 0.1857408881187439, + "learning_rate": 0.0004821999368218292, + "loss": 1.9054, + "step": 1567 + }, + { + "epoch": 0.153125, + "grad_norm": 0.23223114013671875, + "learning_rate": 0.0004821722049810002, + "loss": 1.9118, + "step": 1568 + }, + { + "epoch": 0.15322265625, + "grad_norm": 0.25117412209510803, + "learning_rate": 0.0004821444524455034, + "loss": 1.8889, + "step": 1569 + }, + { + "epoch": 0.1533203125, + "grad_norm": 0.23292681574821472, + "learning_rate": 0.00048211667921811084, + "loss": 1.9495, + "step": 1570 + }, + { + "epoch": 0.15341796875, + "grad_norm": 0.24508163332939148, + "learning_rate": 0.0004820888853015969, + "loss": 1.922, + "step": 1571 + }, + { + "epoch": 0.153515625, + "grad_norm": 0.26448142528533936, + "learning_rate": 0.000482061070698738, + "loss": 1.9811, + "step": 1572 + }, + { + "epoch": 0.15361328125, + "grad_norm": 0.24274107813835144, + "learning_rate": 0.0004820332354123125, + "loss": 1.9134, + "step": 1573 + }, + { + "epoch": 0.1537109375, + "grad_norm": 0.2016661912202835, + "learning_rate": 0.00048200537944510097, + "loss": 1.9151, + "step": 1574 + }, + { + "epoch": 0.15380859375, + "grad_norm": 0.21885213255882263, + "learning_rate": 0.00048197750279988595, + "loss": 1.9613, + "step": 1575 + }, + { + "epoch": 0.15390625, + "grad_norm": 0.2705588936805725, + "learning_rate": 0.000481949605479452, + "loss": 1.9535, + "step": 1576 + }, + { + "epoch": 0.15400390625, + "grad_norm": 0.31828904151916504, + "learning_rate": 0.00048192168748658595, + "loss": 1.9592, + "step": 1577 + }, + { + "epoch": 0.1541015625, + "grad_norm": 0.38848698139190674, + "learning_rate": 0.0004818937488240764, + "loss": 1.9027, + "step": 1578 + }, + { + "epoch": 0.15419921875, + "grad_norm": 0.2857527732849121, + "learning_rate": 0.00048186578949471434, + "loss": 1.9888, + "step": 1579 + }, + { + "epoch": 0.154296875, + "grad_norm": 0.30096855759620667, + "learning_rate": 0.0004818378095012925, + "loss": 1.9199, + "step": 1580 + }, + { + "epoch": 0.15439453125, + "grad_norm": 0.3158320486545563, + "learning_rate": 0.00048180980884660594, + "loss": 1.9081, + "step": 1581 + }, + { + "epoch": 0.1544921875, + "grad_norm": 0.22765566408634186, + "learning_rate": 0.00048178178753345157, + "loss": 1.9181, + "step": 1582 + }, + { + "epoch": 0.15458984375, + "grad_norm": 0.28936147689819336, + "learning_rate": 0.0004817537455646286, + "loss": 1.9424, + "step": 1583 + }, + { + "epoch": 0.1546875, + "grad_norm": 0.3088884651660919, + "learning_rate": 0.00048172568294293816, + "loss": 1.9088, + "step": 1584 + }, + { + "epoch": 0.15478515625, + "grad_norm": 0.2758021950721741, + "learning_rate": 0.00048169759967118343, + "loss": 1.9298, + "step": 1585 + }, + { + "epoch": 0.1548828125, + "grad_norm": 0.2999780774116516, + "learning_rate": 0.0004816694957521696, + "loss": 1.9335, + "step": 1586 + }, + { + "epoch": 0.15498046875, + "grad_norm": 0.22048184275627136, + "learning_rate": 0.0004816413711887041, + "loss": 1.8843, + "step": 1587 + }, + { + "epoch": 0.155078125, + "grad_norm": 0.26630088686943054, + "learning_rate": 0.0004816132259835963, + "loss": 1.9073, + "step": 1588 + }, + { + "epoch": 0.15517578125, + "grad_norm": 0.2615686058998108, + "learning_rate": 0.0004815850601396577, + "loss": 1.9313, + "step": 1589 + }, + { + "epoch": 0.1552734375, + "grad_norm": 0.2473047822713852, + "learning_rate": 0.00048155687365970167, + "loss": 1.8878, + "step": 1590 + }, + { + "epoch": 0.15537109375, + "grad_norm": 0.3328106999397278, + "learning_rate": 0.000481528666546544, + "loss": 1.9424, + "step": 1591 + }, + { + "epoch": 0.15546875, + "grad_norm": 0.26094701886177063, + "learning_rate": 0.0004815004388030022, + "loss": 1.9623, + "step": 1592 + }, + { + "epoch": 0.15556640625, + "grad_norm": 0.3564736247062683, + "learning_rate": 0.0004814721904318959, + "loss": 1.8966, + "step": 1593 + }, + { + "epoch": 0.1556640625, + "grad_norm": 0.3765439987182617, + "learning_rate": 0.000481443921436047, + "loss": 1.9189, + "step": 1594 + }, + { + "epoch": 0.15576171875, + "grad_norm": 0.36775973439216614, + "learning_rate": 0.0004814156318182793, + "loss": 1.9427, + "step": 1595 + }, + { + "epoch": 0.155859375, + "grad_norm": 0.3822444975376129, + "learning_rate": 0.00048138732158141867, + "loss": 1.9329, + "step": 1596 + }, + { + "epoch": 0.15595703125, + "grad_norm": 0.26008525490760803, + "learning_rate": 0.00048135899072829295, + "loss": 1.928, + "step": 1597 + }, + { + "epoch": 0.1560546875, + "grad_norm": 0.32837969064712524, + "learning_rate": 0.0004813306392617324, + "loss": 1.9409, + "step": 1598 + }, + { + "epoch": 0.15615234375, + "grad_norm": 0.30874279141426086, + "learning_rate": 0.0004813022671845687, + "loss": 1.9378, + "step": 1599 + }, + { + "epoch": 0.15625, + "grad_norm": 0.2988525629043579, + "learning_rate": 0.00048127387449963633, + "loss": 1.9091, + "step": 1600 + }, + { + "epoch": 0.15634765625, + "grad_norm": 0.4251779615879059, + "learning_rate": 0.0004812454612097713, + "loss": 1.9693, + "step": 1601 + }, + { + "epoch": 0.1564453125, + "grad_norm": 0.2827276885509491, + "learning_rate": 0.00048121702731781184, + "loss": 1.9177, + "step": 1602 + }, + { + "epoch": 0.15654296875, + "grad_norm": 0.24983742833137512, + "learning_rate": 0.00048118857282659834, + "loss": 1.9235, + "step": 1603 + }, + { + "epoch": 0.156640625, + "grad_norm": 0.25737494230270386, + "learning_rate": 0.000481160097738973, + "loss": 1.9678, + "step": 1604 + }, + { + "epoch": 0.15673828125, + "grad_norm": 0.38471952080726624, + "learning_rate": 0.00048113160205778035, + "loss": 1.9403, + "step": 1605 + }, + { + "epoch": 0.1568359375, + "grad_norm": 0.30553486943244934, + "learning_rate": 0.0004811030857858667, + "loss": 1.939, + "step": 1606 + }, + { + "epoch": 0.15693359375, + "grad_norm": 0.3170030117034912, + "learning_rate": 0.0004810745489260808, + "loss": 1.9331, + "step": 1607 + }, + { + "epoch": 0.15703125, + "grad_norm": 0.3244478404521942, + "learning_rate": 0.0004810459914812731, + "loss": 1.8986, + "step": 1608 + }, + { + "epoch": 0.15712890625, + "grad_norm": 0.24035608768463135, + "learning_rate": 0.0004810174134542963, + "loss": 1.9542, + "step": 1609 + }, + { + "epoch": 0.1572265625, + "grad_norm": 0.29910752177238464, + "learning_rate": 0.00048098881484800505, + "loss": 1.8521, + "step": 1610 + }, + { + "epoch": 0.15732421875, + "grad_norm": 0.2832294702529907, + "learning_rate": 0.000480960195665256, + "loss": 1.9567, + "step": 1611 + }, + { + "epoch": 0.157421875, + "grad_norm": 0.3093940317630768, + "learning_rate": 0.0004809315559089081, + "loss": 1.9332, + "step": 1612 + }, + { + "epoch": 0.15751953125, + "grad_norm": 0.2190847098827362, + "learning_rate": 0.00048090289558182215, + "loss": 1.8979, + "step": 1613 + }, + { + "epoch": 0.1576171875, + "grad_norm": 0.2461078017950058, + "learning_rate": 0.00048087421468686107, + "loss": 1.9032, + "step": 1614 + }, + { + "epoch": 0.15771484375, + "grad_norm": 0.2852519750595093, + "learning_rate": 0.0004808455132268898, + "loss": 1.9457, + "step": 1615 + }, + { + "epoch": 0.1578125, + "grad_norm": 0.23334845900535583, + "learning_rate": 0.00048081679120477543, + "loss": 1.8718, + "step": 1616 + }, + { + "epoch": 0.15791015625, + "grad_norm": 0.2714453637599945, + "learning_rate": 0.000480788048623387, + "loss": 1.9212, + "step": 1617 + }, + { + "epoch": 0.1580078125, + "grad_norm": 0.2445923238992691, + "learning_rate": 0.0004807592854855955, + "loss": 1.9322, + "step": 1618 + }, + { + "epoch": 0.15810546875, + "grad_norm": 0.2522878050804138, + "learning_rate": 0.0004807305017942744, + "loss": 1.939, + "step": 1619 + }, + { + "epoch": 0.158203125, + "grad_norm": 0.28525444865226746, + "learning_rate": 0.0004807016975522987, + "loss": 1.929, + "step": 1620 + }, + { + "epoch": 0.15830078125, + "grad_norm": 0.22004222869873047, + "learning_rate": 0.0004806728727625458, + "loss": 1.9431, + "step": 1621 + }, + { + "epoch": 0.1583984375, + "grad_norm": 0.20370328426361084, + "learning_rate": 0.000480644027427895, + "loss": 1.9063, + "step": 1622 + }, + { + "epoch": 0.15849609375, + "grad_norm": 0.21263106167316437, + "learning_rate": 0.00048061516155122763, + "loss": 1.9098, + "step": 1623 + }, + { + "epoch": 0.15859375, + "grad_norm": 0.25402888655662537, + "learning_rate": 0.0004805862751354273, + "loss": 1.8932, + "step": 1624 + }, + { + "epoch": 0.15869140625, + "grad_norm": 0.21452878415584564, + "learning_rate": 0.0004805573681833794, + "loss": 1.9674, + "step": 1625 + }, + { + "epoch": 0.1587890625, + "grad_norm": 0.2378520965576172, + "learning_rate": 0.0004805284406979715, + "loss": 1.8841, + "step": 1626 + }, + { + "epoch": 0.15888671875, + "grad_norm": 0.20481383800506592, + "learning_rate": 0.0004804994926820932, + "loss": 1.8836, + "step": 1627 + }, + { + "epoch": 0.158984375, + "grad_norm": 0.19195064902305603, + "learning_rate": 0.00048047052413863607, + "loss": 1.9694, + "step": 1628 + }, + { + "epoch": 0.15908203125, + "grad_norm": 0.228655144572258, + "learning_rate": 0.0004804415350704939, + "loss": 1.9323, + "step": 1629 + }, + { + "epoch": 0.1591796875, + "grad_norm": 0.2156544327735901, + "learning_rate": 0.0004804125254805625, + "loss": 1.9236, + "step": 1630 + }, + { + "epoch": 0.15927734375, + "grad_norm": 0.24159860610961914, + "learning_rate": 0.00048038349537173953, + "loss": 1.9611, + "step": 1631 + }, + { + "epoch": 0.159375, + "grad_norm": 0.21641957759857178, + "learning_rate": 0.0004803544447469249, + "loss": 1.9237, + "step": 1632 + }, + { + "epoch": 0.15947265625, + "grad_norm": 0.21140074729919434, + "learning_rate": 0.0004803253736090205, + "loss": 1.9297, + "step": 1633 + }, + { + "epoch": 0.1595703125, + "grad_norm": 0.23588809370994568, + "learning_rate": 0.00048029628196093047, + "loss": 1.8995, + "step": 1634 + }, + { + "epoch": 0.15966796875, + "grad_norm": 0.2083786576986313, + "learning_rate": 0.00048026716980556044, + "loss": 1.9154, + "step": 1635 + }, + { + "epoch": 0.159765625, + "grad_norm": 0.21067486703395844, + "learning_rate": 0.00048023803714581867, + "loss": 1.9032, + "step": 1636 + }, + { + "epoch": 0.15986328125, + "grad_norm": 0.19560249149799347, + "learning_rate": 0.0004802088839846153, + "loss": 1.9352, + "step": 1637 + }, + { + "epoch": 0.1599609375, + "grad_norm": 0.21477989852428436, + "learning_rate": 0.0004801797103248624, + "loss": 1.9192, + "step": 1638 + }, + { + "epoch": 0.16005859375, + "grad_norm": 0.27302122116088867, + "learning_rate": 0.0004801505161694741, + "loss": 1.9177, + "step": 1639 + }, + { + "epoch": 0.16015625, + "grad_norm": 0.251449316740036, + "learning_rate": 0.00048012130152136676, + "loss": 1.9383, + "step": 1640 + }, + { + "epoch": 0.16025390625, + "grad_norm": 0.29725855588912964, + "learning_rate": 0.00048009206638345857, + "loss": 1.9027, + "step": 1641 + }, + { + "epoch": 0.1603515625, + "grad_norm": 0.3364620804786682, + "learning_rate": 0.00048006281075866996, + "loss": 1.8999, + "step": 1642 + }, + { + "epoch": 0.16044921875, + "grad_norm": 0.33240416646003723, + "learning_rate": 0.00048003353464992323, + "loss": 1.9023, + "step": 1643 + }, + { + "epoch": 0.160546875, + "grad_norm": 0.29456380009651184, + "learning_rate": 0.0004800042380601428, + "loss": 1.8874, + "step": 1644 + }, + { + "epoch": 0.16064453125, + "grad_norm": 0.20352846384048462, + "learning_rate": 0.00047997492099225515, + "loss": 1.8813, + "step": 1645 + }, + { + "epoch": 0.1607421875, + "grad_norm": 0.2666397988796234, + "learning_rate": 0.0004799455834491889, + "loss": 1.9582, + "step": 1646 + }, + { + "epoch": 0.16083984375, + "grad_norm": 0.34426215291023254, + "learning_rate": 0.0004799162254338744, + "loss": 1.9372, + "step": 1647 + }, + { + "epoch": 0.1609375, + "grad_norm": 0.3887472450733185, + "learning_rate": 0.00047988684694924447, + "loss": 1.9381, + "step": 1648 + }, + { + "epoch": 0.16103515625, + "grad_norm": 0.3413529694080353, + "learning_rate": 0.0004798574479982336, + "loss": 1.8966, + "step": 1649 + }, + { + "epoch": 0.1611328125, + "grad_norm": 0.2743453085422516, + "learning_rate": 0.00047982802858377853, + "loss": 1.9228, + "step": 1650 + }, + { + "epoch": 0.16123046875, + "grad_norm": 0.29950737953186035, + "learning_rate": 0.00047979858870881805, + "loss": 1.9204, + "step": 1651 + }, + { + "epoch": 0.161328125, + "grad_norm": 0.273075133562088, + "learning_rate": 0.00047976912837629295, + "loss": 1.8683, + "step": 1652 + }, + { + "epoch": 0.16142578125, + "grad_norm": 0.2547522485256195, + "learning_rate": 0.00047973964758914596, + "loss": 1.9275, + "step": 1653 + }, + { + "epoch": 0.1615234375, + "grad_norm": 0.2716279923915863, + "learning_rate": 0.00047971014635032205, + "loss": 1.8917, + "step": 1654 + }, + { + "epoch": 0.16162109375, + "grad_norm": 0.28852027654647827, + "learning_rate": 0.00047968062466276803, + "loss": 1.9418, + "step": 1655 + }, + { + "epoch": 0.16171875, + "grad_norm": 0.19788789749145508, + "learning_rate": 0.00047965108252943293, + "loss": 1.922, + "step": 1656 + }, + { + "epoch": 0.16181640625, + "grad_norm": 0.33596405386924744, + "learning_rate": 0.0004796215199532678, + "loss": 1.9291, + "step": 1657 + }, + { + "epoch": 0.1619140625, + "grad_norm": 0.32864540815353394, + "learning_rate": 0.00047959193693722553, + "loss": 1.9793, + "step": 1658 + }, + { + "epoch": 0.16201171875, + "grad_norm": 0.25629329681396484, + "learning_rate": 0.0004795623334842613, + "loss": 1.9453, + "step": 1659 + }, + { + "epoch": 0.162109375, + "grad_norm": 0.2877620458602905, + "learning_rate": 0.0004795327095973322, + "loss": 1.9225, + "step": 1660 + }, + { + "epoch": 0.16220703125, + "grad_norm": 0.24507132172584534, + "learning_rate": 0.00047950306527939744, + "loss": 1.8917, + "step": 1661 + }, + { + "epoch": 0.1623046875, + "grad_norm": 0.2875724732875824, + "learning_rate": 0.0004794734005334182, + "loss": 1.8819, + "step": 1662 + }, + { + "epoch": 0.16240234375, + "grad_norm": 0.2515362501144409, + "learning_rate": 0.0004794437153623577, + "loss": 1.9232, + "step": 1663 + }, + { + "epoch": 0.1625, + "grad_norm": 0.27037444710731506, + "learning_rate": 0.00047941400976918123, + "loss": 1.9567, + "step": 1664 + }, + { + "epoch": 0.16259765625, + "grad_norm": 0.29190242290496826, + "learning_rate": 0.0004793842837568562, + "loss": 1.9266, + "step": 1665 + }, + { + "epoch": 0.1626953125, + "grad_norm": 0.26770535111427307, + "learning_rate": 0.0004793545373283518, + "loss": 1.9163, + "step": 1666 + }, + { + "epoch": 0.16279296875, + "grad_norm": 0.25330448150634766, + "learning_rate": 0.0004793247704866396, + "loss": 1.9182, + "step": 1667 + }, + { + "epoch": 0.162890625, + "grad_norm": 0.27300313115119934, + "learning_rate": 0.00047929498323469295, + "loss": 1.9592, + "step": 1668 + }, + { + "epoch": 0.16298828125, + "grad_norm": 0.21413999795913696, + "learning_rate": 0.00047926517557548746, + "loss": 1.919, + "step": 1669 + }, + { + "epoch": 0.1630859375, + "grad_norm": 0.27660825848579407, + "learning_rate": 0.0004792353475120004, + "loss": 1.9176, + "step": 1670 + }, + { + "epoch": 0.16318359375, + "grad_norm": 0.2682628631591797, + "learning_rate": 0.00047920549904721156, + "loss": 1.9381, + "step": 1671 + }, + { + "epoch": 0.16328125, + "grad_norm": 0.2719988226890564, + "learning_rate": 0.00047917563018410247, + "loss": 1.9137, + "step": 1672 + }, + { + "epoch": 0.16337890625, + "grad_norm": 0.23097728192806244, + "learning_rate": 0.0004791457409256568, + "loss": 1.9128, + "step": 1673 + }, + { + "epoch": 0.1634765625, + "grad_norm": 0.27505284547805786, + "learning_rate": 0.00047911583127486015, + "loss": 1.9026, + "step": 1674 + }, + { + "epoch": 0.16357421875, + "grad_norm": 0.31115567684173584, + "learning_rate": 0.0004790859012347002, + "loss": 1.9438, + "step": 1675 + }, + { + "epoch": 0.163671875, + "grad_norm": 0.2766979932785034, + "learning_rate": 0.0004790559508081668, + "loss": 1.9389, + "step": 1676 + }, + { + "epoch": 0.16376953125, + "grad_norm": 0.25364476442337036, + "learning_rate": 0.0004790259799982517, + "loss": 1.9621, + "step": 1677 + }, + { + "epoch": 0.1638671875, + "grad_norm": 0.2380513697862625, + "learning_rate": 0.0004789959888079487, + "loss": 1.8891, + "step": 1678 + }, + { + "epoch": 0.16396484375, + "grad_norm": 0.21375243365764618, + "learning_rate": 0.0004789659772402536, + "loss": 1.8864, + "step": 1679 + }, + { + "epoch": 0.1640625, + "grad_norm": 0.2307070791721344, + "learning_rate": 0.00047893594529816445, + "loss": 1.8975, + "step": 1680 + }, + { + "epoch": 0.16416015625, + "grad_norm": 0.20362551510334015, + "learning_rate": 0.000478905892984681, + "loss": 1.9489, + "step": 1681 + }, + { + "epoch": 0.1642578125, + "grad_norm": 0.23438306152820587, + "learning_rate": 0.00047887582030280536, + "loss": 1.9874, + "step": 1682 + }, + { + "epoch": 0.16435546875, + "grad_norm": 0.19733543694019318, + "learning_rate": 0.00047884572725554137, + "loss": 1.9147, + "step": 1683 + }, + { + "epoch": 0.164453125, + "grad_norm": 0.2748172879219055, + "learning_rate": 0.0004788156138458952, + "loss": 1.863, + "step": 1684 + }, + { + "epoch": 0.16455078125, + "grad_norm": 0.26911303400993347, + "learning_rate": 0.00047878548007687476, + "loss": 1.8852, + "step": 1685 + }, + { + "epoch": 0.1646484375, + "grad_norm": 0.20826198160648346, + "learning_rate": 0.0004787553259514903, + "loss": 1.933, + "step": 1686 + }, + { + "epoch": 0.16474609375, + "grad_norm": 0.3136061429977417, + "learning_rate": 0.00047872515147275393, + "loss": 1.9264, + "step": 1687 + }, + { + "epoch": 0.16484375, + "grad_norm": 0.22931191325187683, + "learning_rate": 0.0004786949566436797, + "loss": 1.9175, + "step": 1688 + }, + { + "epoch": 0.16494140625, + "grad_norm": 0.24129685759544373, + "learning_rate": 0.0004786647414672839, + "loss": 1.9114, + "step": 1689 + }, + { + "epoch": 0.1650390625, + "grad_norm": 0.2148384302854538, + "learning_rate": 0.00047863450594658473, + "loss": 1.9189, + "step": 1690 + }, + { + "epoch": 0.16513671875, + "grad_norm": 0.24560751020908356, + "learning_rate": 0.0004786042500846025, + "loss": 1.9459, + "step": 1691 + }, + { + "epoch": 0.165234375, + "grad_norm": 0.2328975796699524, + "learning_rate": 0.0004785739738843594, + "loss": 1.9037, + "step": 1692 + }, + { + "epoch": 0.16533203125, + "grad_norm": 0.19054313004016876, + "learning_rate": 0.00047854367734887984, + "loss": 1.9035, + "step": 1693 + }, + { + "epoch": 0.1654296875, + "grad_norm": 0.19610174000263214, + "learning_rate": 0.00047851336048119016, + "loss": 1.8709, + "step": 1694 + }, + { + "epoch": 0.16552734375, + "grad_norm": 0.22446346282958984, + "learning_rate": 0.0004784830232843186, + "loss": 1.8866, + "step": 1695 + }, + { + "epoch": 0.165625, + "grad_norm": 0.22055386006832123, + "learning_rate": 0.0004784526657612959, + "loss": 1.9104, + "step": 1696 + }, + { + "epoch": 0.16572265625, + "grad_norm": 0.19776830077171326, + "learning_rate": 0.0004784222879151542, + "loss": 1.9302, + "step": 1697 + }, + { + "epoch": 0.1658203125, + "grad_norm": 0.19348515570163727, + "learning_rate": 0.0004783918897489282, + "loss": 1.944, + "step": 1698 + }, + { + "epoch": 0.16591796875, + "grad_norm": 0.21035566926002502, + "learning_rate": 0.0004783614712656542, + "loss": 1.9697, + "step": 1699 + }, + { + "epoch": 0.166015625, + "grad_norm": 0.20167845487594604, + "learning_rate": 0.0004783310324683709, + "loss": 1.9087, + "step": 1700 + }, + { + "epoch": 0.16611328125, + "grad_norm": 0.24482861161231995, + "learning_rate": 0.0004783005733601188, + "loss": 1.9316, + "step": 1701 + }, + { + "epoch": 0.1662109375, + "grad_norm": 0.2270205020904541, + "learning_rate": 0.00047827009394394045, + "loss": 1.885, + "step": 1702 + }, + { + "epoch": 0.16630859375, + "grad_norm": 0.2608849108219147, + "learning_rate": 0.0004782395942228806, + "loss": 1.9639, + "step": 1703 + }, + { + "epoch": 0.16640625, + "grad_norm": 0.29610663652420044, + "learning_rate": 0.0004782090741999858, + "loss": 1.9248, + "step": 1704 + }, + { + "epoch": 0.16650390625, + "grad_norm": 0.44481170177459717, + "learning_rate": 0.0004781785338783047, + "loss": 1.9352, + "step": 1705 + }, + { + "epoch": 0.1666015625, + "grad_norm": 0.4372797906398773, + "learning_rate": 0.0004781479732608881, + "loss": 1.9557, + "step": 1706 + }, + { + "epoch": 0.16669921875, + "grad_norm": 0.2859649956226349, + "learning_rate": 0.00047811739235078876, + "loss": 1.9753, + "step": 1707 + }, + { + "epoch": 0.166796875, + "grad_norm": 0.26134198904037476, + "learning_rate": 0.00047808679115106135, + "loss": 1.9513, + "step": 1708 + }, + { + "epoch": 0.16689453125, + "grad_norm": 0.30859240889549255, + "learning_rate": 0.0004780561696647627, + "loss": 1.9084, + "step": 1709 + }, + { + "epoch": 0.1669921875, + "grad_norm": 0.33629563450813293, + "learning_rate": 0.0004780255278949516, + "loss": 1.928, + "step": 1710 + }, + { + "epoch": 0.16708984375, + "grad_norm": 0.2672228217124939, + "learning_rate": 0.0004779948658446889, + "loss": 1.9033, + "step": 1711 + }, + { + "epoch": 0.1671875, + "grad_norm": 0.2845448851585388, + "learning_rate": 0.00047796418351703744, + "loss": 1.8892, + "step": 1712 + }, + { + "epoch": 0.16728515625, + "grad_norm": 0.236465185880661, + "learning_rate": 0.0004779334809150622, + "loss": 1.9363, + "step": 1713 + }, + { + "epoch": 0.1673828125, + "grad_norm": 0.2997010052204132, + "learning_rate": 0.00047790275804183004, + "loss": 1.8917, + "step": 1714 + }, + { + "epoch": 0.16748046875, + "grad_norm": 0.3848778307437897, + "learning_rate": 0.00047787201490040985, + "loss": 1.959, + "step": 1715 + }, + { + "epoch": 0.167578125, + "grad_norm": 0.2715388834476471, + "learning_rate": 0.00047784125149387277, + "loss": 1.9463, + "step": 1716 + }, + { + "epoch": 0.16767578125, + "grad_norm": 0.3423638343811035, + "learning_rate": 0.0004778104678252916, + "loss": 1.9393, + "step": 1717 + }, + { + "epoch": 0.1677734375, + "grad_norm": 0.3216153383255005, + "learning_rate": 0.0004777796638977414, + "loss": 1.9688, + "step": 1718 + }, + { + "epoch": 0.16787109375, + "grad_norm": 0.24108828604221344, + "learning_rate": 0.00047774883971429935, + "loss": 1.8822, + "step": 1719 + }, + { + "epoch": 0.16796875, + "grad_norm": 0.3516473174095154, + "learning_rate": 0.0004777179952780443, + "loss": 1.912, + "step": 1720 + }, + { + "epoch": 0.16806640625, + "grad_norm": 0.2251645028591156, + "learning_rate": 0.0004776871305920575, + "loss": 1.8989, + "step": 1721 + }, + { + "epoch": 0.1681640625, + "grad_norm": 0.282797634601593, + "learning_rate": 0.000477656245659422, + "loss": 1.8483, + "step": 1722 + }, + { + "epoch": 0.16826171875, + "grad_norm": 0.3045579195022583, + "learning_rate": 0.0004776253404832229, + "loss": 1.8934, + "step": 1723 + }, + { + "epoch": 0.168359375, + "grad_norm": 0.27416738867759705, + "learning_rate": 0.00047759441506654745, + "loss": 1.924, + "step": 1724 + }, + { + "epoch": 0.16845703125, + "grad_norm": 0.2600431442260742, + "learning_rate": 0.0004775634694124847, + "loss": 1.9507, + "step": 1725 + }, + { + "epoch": 0.1685546875, + "grad_norm": 0.22953875362873077, + "learning_rate": 0.0004775325035241259, + "loss": 1.9062, + "step": 1726 + }, + { + "epoch": 0.16865234375, + "grad_norm": 0.20656803250312805, + "learning_rate": 0.00047750151740456426, + "loss": 1.9312, + "step": 1727 + }, + { + "epoch": 0.16875, + "grad_norm": 0.21857194602489471, + "learning_rate": 0.0004774705110568951, + "loss": 1.931, + "step": 1728 + }, + { + "epoch": 0.16884765625, + "grad_norm": 0.2585659325122833, + "learning_rate": 0.0004774394844842155, + "loss": 1.9149, + "step": 1729 + }, + { + "epoch": 0.1689453125, + "grad_norm": 0.23686473071575165, + "learning_rate": 0.000477408437689625, + "loss": 1.9316, + "step": 1730 + }, + { + "epoch": 0.16904296875, + "grad_norm": 0.21281464397907257, + "learning_rate": 0.0004773773706762247, + "loss": 1.9278, + "step": 1731 + }, + { + "epoch": 0.169140625, + "grad_norm": 0.2731045186519623, + "learning_rate": 0.00047734628344711796, + "loss": 1.9088, + "step": 1732 + }, + { + "epoch": 0.16923828125, + "grad_norm": 0.21235254406929016, + "learning_rate": 0.00047731517600541014, + "loss": 1.9418, + "step": 1733 + }, + { + "epoch": 0.1693359375, + "grad_norm": 0.2536609172821045, + "learning_rate": 0.0004772840483542085, + "loss": 1.9651, + "step": 1734 + }, + { + "epoch": 0.16943359375, + "grad_norm": 0.22243930399417877, + "learning_rate": 0.00047725290049662264, + "loss": 1.9158, + "step": 1735 + }, + { + "epoch": 0.16953125, + "grad_norm": 0.24414226412773132, + "learning_rate": 0.00047722173243576374, + "loss": 1.9167, + "step": 1736 + }, + { + "epoch": 0.16962890625, + "grad_norm": 0.25493788719177246, + "learning_rate": 0.0004771905441747453, + "loss": 1.9244, + "step": 1737 + }, + { + "epoch": 0.1697265625, + "grad_norm": 0.21703946590423584, + "learning_rate": 0.0004771593357166827, + "loss": 1.8882, + "step": 1738 + }, + { + "epoch": 0.16982421875, + "grad_norm": 0.27353623509407043, + "learning_rate": 0.00047712810706469354, + "loss": 1.9134, + "step": 1739 + }, + { + "epoch": 0.169921875, + "grad_norm": 0.2163049578666687, + "learning_rate": 0.0004770968582218971, + "loss": 1.9226, + "step": 1740 + }, + { + "epoch": 0.17001953125, + "grad_norm": 0.1948620080947876, + "learning_rate": 0.000477065589191415, + "loss": 1.8663, + "step": 1741 + }, + { + "epoch": 0.1701171875, + "grad_norm": 0.22654829919338226, + "learning_rate": 0.0004770342999763706, + "loss": 1.9327, + "step": 1742 + }, + { + "epoch": 0.17021484375, + "grad_norm": 0.24024905264377594, + "learning_rate": 0.00047700299057988957, + "loss": 1.9581, + "step": 1743 + }, + { + "epoch": 0.1703125, + "grad_norm": 0.2727830111980438, + "learning_rate": 0.0004769716610050993, + "loss": 1.9158, + "step": 1744 + }, + { + "epoch": 0.17041015625, + "grad_norm": 0.2848643362522125, + "learning_rate": 0.0004769403112551295, + "loss": 1.8943, + "step": 1745 + }, + { + "epoch": 0.1705078125, + "grad_norm": 0.29621654748916626, + "learning_rate": 0.00047690894133311157, + "loss": 1.9182, + "step": 1746 + }, + { + "epoch": 0.17060546875, + "grad_norm": 0.317327618598938, + "learning_rate": 0.00047687755124217914, + "loss": 1.9624, + "step": 1747 + }, + { + "epoch": 0.170703125, + "grad_norm": 0.3084838390350342, + "learning_rate": 0.00047684614098546783, + "loss": 1.919, + "step": 1748 + }, + { + "epoch": 0.17080078125, + "grad_norm": 0.26487037539482117, + "learning_rate": 0.00047681471056611527, + "loss": 1.9005, + "step": 1749 + }, + { + "epoch": 0.1708984375, + "grad_norm": 0.2641773223876953, + "learning_rate": 0.000476783259987261, + "loss": 1.9085, + "step": 1750 + }, + { + "epoch": 0.17099609375, + "grad_norm": 0.30503541231155396, + "learning_rate": 0.00047675178925204667, + "loss": 1.9148, + "step": 1751 + }, + { + "epoch": 0.17109375, + "grad_norm": 0.2899084985256195, + "learning_rate": 0.00047672029836361596, + "loss": 1.9369, + "step": 1752 + }, + { + "epoch": 0.17119140625, + "grad_norm": 0.3330906629562378, + "learning_rate": 0.00047668878732511457, + "loss": 1.8529, + "step": 1753 + }, + { + "epoch": 0.1712890625, + "grad_norm": 0.32898756861686707, + "learning_rate": 0.00047665725613969015, + "loss": 1.8918, + "step": 1754 + }, + { + "epoch": 0.17138671875, + "grad_norm": 0.21951165795326233, + "learning_rate": 0.0004766257048104923, + "loss": 1.8868, + "step": 1755 + }, + { + "epoch": 0.171484375, + "grad_norm": 0.20533274114131927, + "learning_rate": 0.00047659413334067276, + "loss": 1.8993, + "step": 1756 + }, + { + "epoch": 0.17158203125, + "grad_norm": 0.19753561913967133, + "learning_rate": 0.0004765625417333853, + "loss": 1.913, + "step": 1757 + }, + { + "epoch": 0.1716796875, + "grad_norm": 0.20068642497062683, + "learning_rate": 0.0004765309299917856, + "loss": 1.8639, + "step": 1758 + }, + { + "epoch": 0.17177734375, + "grad_norm": 0.22603540122509003, + "learning_rate": 0.00047649929811903143, + "loss": 1.9014, + "step": 1759 + }, + { + "epoch": 0.171875, + "grad_norm": 0.18157346546649933, + "learning_rate": 0.00047646764611828247, + "loss": 1.8956, + "step": 1760 + }, + { + "epoch": 0.17197265625, + "grad_norm": 0.21239453554153442, + "learning_rate": 0.0004764359739927005, + "loss": 1.9636, + "step": 1761 + }, + { + "epoch": 0.1720703125, + "grad_norm": 0.25655651092529297, + "learning_rate": 0.0004764042817454494, + "loss": 1.8901, + "step": 1762 + }, + { + "epoch": 0.17216796875, + "grad_norm": 0.20679731667041779, + "learning_rate": 0.0004763725693796947, + "loss": 1.8908, + "step": 1763 + }, + { + "epoch": 0.172265625, + "grad_norm": 0.20754674077033997, + "learning_rate": 0.0004763408368986043, + "loss": 1.9445, + "step": 1764 + }, + { + "epoch": 0.17236328125, + "grad_norm": 0.18912455439567566, + "learning_rate": 0.0004763090843053481, + "loss": 1.9176, + "step": 1765 + }, + { + "epoch": 0.1724609375, + "grad_norm": 0.20957696437835693, + "learning_rate": 0.0004762773116030978, + "loss": 1.9206, + "step": 1766 + }, + { + "epoch": 0.17255859375, + "grad_norm": 0.1931106299161911, + "learning_rate": 0.0004762455187950272, + "loss": 1.9051, + "step": 1767 + }, + { + "epoch": 0.17265625, + "grad_norm": 0.218103289604187, + "learning_rate": 0.00047621370588431217, + "loss": 1.9326, + "step": 1768 + }, + { + "epoch": 0.17275390625, + "grad_norm": 0.20649409294128418, + "learning_rate": 0.0004761818728741305, + "loss": 1.9156, + "step": 1769 + }, + { + "epoch": 0.1728515625, + "grad_norm": 0.2224937528371811, + "learning_rate": 0.000476150019767662, + "loss": 1.9181, + "step": 1770 + }, + { + "epoch": 0.17294921875, + "grad_norm": 0.30575814843177795, + "learning_rate": 0.00047611814656808865, + "loss": 1.9276, + "step": 1771 + }, + { + "epoch": 0.173046875, + "grad_norm": 0.34954264760017395, + "learning_rate": 0.0004760862532785942, + "loss": 1.8818, + "step": 1772 + }, + { + "epoch": 0.17314453125, + "grad_norm": 0.2726197838783264, + "learning_rate": 0.0004760543399023644, + "loss": 1.9279, + "step": 1773 + }, + { + "epoch": 0.1732421875, + "grad_norm": 0.2664092779159546, + "learning_rate": 0.00047602240644258726, + "loss": 1.9055, + "step": 1774 + }, + { + "epoch": 0.17333984375, + "grad_norm": 0.22661474347114563, + "learning_rate": 0.00047599045290245277, + "loss": 1.9406, + "step": 1775 + }, + { + "epoch": 0.1734375, + "grad_norm": 0.20825809240341187, + "learning_rate": 0.00047595847928515243, + "loss": 1.9189, + "step": 1776 + }, + { + "epoch": 0.17353515625, + "grad_norm": 0.200863778591156, + "learning_rate": 0.0004759264855938805, + "loss": 1.8985, + "step": 1777 + }, + { + "epoch": 0.1736328125, + "grad_norm": 0.18549345433712006, + "learning_rate": 0.00047589447183183254, + "loss": 1.8855, + "step": 1778 + }, + { + "epoch": 0.17373046875, + "grad_norm": 0.19504952430725098, + "learning_rate": 0.00047586243800220673, + "loss": 1.9494, + "step": 1779 + }, + { + "epoch": 0.173828125, + "grad_norm": 0.2109365612268448, + "learning_rate": 0.0004758303841082029, + "loss": 1.9223, + "step": 1780 + }, + { + "epoch": 0.17392578125, + "grad_norm": 0.20636534690856934, + "learning_rate": 0.00047579831015302277, + "loss": 1.8939, + "step": 1781 + }, + { + "epoch": 0.1740234375, + "grad_norm": 0.2035193145275116, + "learning_rate": 0.0004757662161398704, + "loss": 1.933, + "step": 1782 + }, + { + "epoch": 0.17412109375, + "grad_norm": 0.20754744112491608, + "learning_rate": 0.0004757341020719517, + "loss": 1.8626, + "step": 1783 + }, + { + "epoch": 0.17421875, + "grad_norm": 0.29153457283973694, + "learning_rate": 0.00047570196795247455, + "loss": 1.8666, + "step": 1784 + }, + { + "epoch": 0.17431640625, + "grad_norm": 0.36487746238708496, + "learning_rate": 0.00047566981378464884, + "loss": 1.8917, + "step": 1785 + }, + { + "epoch": 0.1744140625, + "grad_norm": 0.32802101969718933, + "learning_rate": 0.0004756376395716865, + "loss": 1.8797, + "step": 1786 + }, + { + "epoch": 0.17451171875, + "grad_norm": 0.2676317095756531, + "learning_rate": 0.0004756054453168015, + "loss": 1.9127, + "step": 1787 + }, + { + "epoch": 0.174609375, + "grad_norm": 0.21236379444599152, + "learning_rate": 0.0004755732310232097, + "loss": 1.8628, + "step": 1788 + }, + { + "epoch": 0.17470703125, + "grad_norm": 0.2659597396850586, + "learning_rate": 0.000475540996694129, + "loss": 1.8725, + "step": 1789 + }, + { + "epoch": 0.1748046875, + "grad_norm": 0.2773102819919586, + "learning_rate": 0.00047550874233277937, + "loss": 1.9394, + "step": 1790 + }, + { + "epoch": 0.17490234375, + "grad_norm": 0.2603963613510132, + "learning_rate": 0.00047547646794238277, + "loss": 1.9164, + "step": 1791 + }, + { + "epoch": 0.175, + "grad_norm": 0.21353478729724884, + "learning_rate": 0.00047544417352616305, + "loss": 1.9091, + "step": 1792 + }, + { + "epoch": 0.17509765625, + "grad_norm": 0.29758280515670776, + "learning_rate": 0.0004754118590873462, + "loss": 1.9137, + "step": 1793 + }, + { + "epoch": 0.1751953125, + "grad_norm": 0.27226370573043823, + "learning_rate": 0.0004753795246291601, + "loss": 1.944, + "step": 1794 + }, + { + "epoch": 0.17529296875, + "grad_norm": 0.3030240833759308, + "learning_rate": 0.0004753471701548347, + "loss": 1.8922, + "step": 1795 + }, + { + "epoch": 0.175390625, + "grad_norm": 0.2778298258781433, + "learning_rate": 0.0004753147956676019, + "loss": 1.9321, + "step": 1796 + }, + { + "epoch": 0.17548828125, + "grad_norm": 0.2760380506515503, + "learning_rate": 0.0004752824011706956, + "loss": 1.9063, + "step": 1797 + }, + { + "epoch": 0.1755859375, + "grad_norm": 0.2793836295604706, + "learning_rate": 0.00047524998666735184, + "loss": 1.9186, + "step": 1798 + }, + { + "epoch": 0.17568359375, + "grad_norm": 0.25465574860572815, + "learning_rate": 0.00047521755216080845, + "loss": 1.9247, + "step": 1799 + }, + { + "epoch": 0.17578125, + "grad_norm": 0.23424822092056274, + "learning_rate": 0.00047518509765430536, + "loss": 1.924, + "step": 1800 + }, + { + "epoch": 0.17587890625, + "grad_norm": 0.25181999802589417, + "learning_rate": 0.00047515262315108456, + "loss": 1.9536, + "step": 1801 + }, + { + "epoch": 0.1759765625, + "grad_norm": 0.2854491174221039, + "learning_rate": 0.00047512012865438976, + "loss": 1.9098, + "step": 1802 + }, + { + "epoch": 0.17607421875, + "grad_norm": 0.3238297998905182, + "learning_rate": 0.0004750876141674671, + "loss": 1.8833, + "step": 1803 + }, + { + "epoch": 0.176171875, + "grad_norm": 0.23803670704364777, + "learning_rate": 0.0004750550796935644, + "loss": 1.9685, + "step": 1804 + }, + { + "epoch": 0.17626953125, + "grad_norm": 0.2600926458835602, + "learning_rate": 0.00047502252523593154, + "loss": 1.913, + "step": 1805 + }, + { + "epoch": 0.1763671875, + "grad_norm": 0.30193689465522766, + "learning_rate": 0.00047498995079782046, + "loss": 1.9453, + "step": 1806 + }, + { + "epoch": 0.17646484375, + "grad_norm": 0.2594965994358063, + "learning_rate": 0.0004749573563824851, + "loss": 1.8932, + "step": 1807 + }, + { + "epoch": 0.1765625, + "grad_norm": 0.25451624393463135, + "learning_rate": 0.0004749247419931812, + "loss": 1.9312, + "step": 1808 + }, + { + "epoch": 0.17666015625, + "grad_norm": 0.2194494754076004, + "learning_rate": 0.00047489210763316674, + "loss": 1.8783, + "step": 1809 + }, + { + "epoch": 0.1767578125, + "grad_norm": 0.2316436916589737, + "learning_rate": 0.00047485945330570173, + "loss": 1.8924, + "step": 1810 + }, + { + "epoch": 0.17685546875, + "grad_norm": 0.2362855076789856, + "learning_rate": 0.00047482677901404785, + "loss": 1.8883, + "step": 1811 + }, + { + "epoch": 0.176953125, + "grad_norm": 0.19301187992095947, + "learning_rate": 0.00047479408476146906, + "loss": 1.8778, + "step": 1812 + }, + { + "epoch": 0.17705078125, + "grad_norm": 0.2641616463661194, + "learning_rate": 0.0004747613705512313, + "loss": 1.8932, + "step": 1813 + }, + { + "epoch": 0.1771484375, + "grad_norm": 0.24684876203536987, + "learning_rate": 0.00047472863638660227, + "loss": 1.8961, + "step": 1814 + }, + { + "epoch": 0.17724609375, + "grad_norm": 0.26180848479270935, + "learning_rate": 0.0004746958822708519, + "loss": 1.8992, + "step": 1815 + }, + { + "epoch": 0.17734375, + "grad_norm": 0.21258847415447235, + "learning_rate": 0.0004746631082072521, + "loss": 1.9513, + "step": 1816 + }, + { + "epoch": 0.17744140625, + "grad_norm": 0.23707158863544464, + "learning_rate": 0.00047463031419907656, + "loss": 1.9015, + "step": 1817 + }, + { + "epoch": 0.1775390625, + "grad_norm": 0.2716965675354004, + "learning_rate": 0.0004745975002496012, + "loss": 1.9023, + "step": 1818 + }, + { + "epoch": 0.17763671875, + "grad_norm": 0.19453154504299164, + "learning_rate": 0.0004745646663621038, + "loss": 1.8883, + "step": 1819 + }, + { + "epoch": 0.177734375, + "grad_norm": 0.28068456053733826, + "learning_rate": 0.00047453181253986437, + "loss": 1.893, + "step": 1820 + }, + { + "epoch": 0.17783203125, + "grad_norm": 0.24416638910770416, + "learning_rate": 0.0004744989387861645, + "loss": 1.9337, + "step": 1821 + }, + { + "epoch": 0.1779296875, + "grad_norm": 0.24311622977256775, + "learning_rate": 0.000474466045104288, + "loss": 1.9139, + "step": 1822 + }, + { + "epoch": 0.17802734375, + "grad_norm": 0.3018793761730194, + "learning_rate": 0.00047443313149752073, + "loss": 1.8941, + "step": 1823 + }, + { + "epoch": 0.178125, + "grad_norm": 0.2428286373615265, + "learning_rate": 0.00047440019796915044, + "loss": 1.9238, + "step": 1824 + }, + { + "epoch": 0.17822265625, + "grad_norm": 0.2676742970943451, + "learning_rate": 0.0004743672445224669, + "loss": 1.8603, + "step": 1825 + }, + { + "epoch": 0.1783203125, + "grad_norm": 0.27228862047195435, + "learning_rate": 0.00047433427116076184, + "loss": 1.9091, + "step": 1826 + }, + { + "epoch": 0.17841796875, + "grad_norm": 0.23455603420734406, + "learning_rate": 0.0004743012778873291, + "loss": 1.8429, + "step": 1827 + }, + { + "epoch": 0.178515625, + "grad_norm": 0.26803067326545715, + "learning_rate": 0.0004742682647054643, + "loss": 1.8386, + "step": 1828 + }, + { + "epoch": 0.17861328125, + "grad_norm": 0.30642521381378174, + "learning_rate": 0.0004742352316184652, + "loss": 1.8997, + "step": 1829 + }, + { + "epoch": 0.1787109375, + "grad_norm": 0.23812299966812134, + "learning_rate": 0.00047420217862963155, + "loss": 1.9538, + "step": 1830 + }, + { + "epoch": 0.17880859375, + "grad_norm": 0.21221864223480225, + "learning_rate": 0.000474169105742265, + "loss": 1.9395, + "step": 1831 + }, + { + "epoch": 0.17890625, + "grad_norm": 0.2294473797082901, + "learning_rate": 0.0004741360129596693, + "loss": 1.9051, + "step": 1832 + }, + { + "epoch": 0.17900390625, + "grad_norm": 0.27911970019340515, + "learning_rate": 0.00047410290028515004, + "loss": 1.9032, + "step": 1833 + }, + { + "epoch": 0.1791015625, + "grad_norm": 0.3017365336418152, + "learning_rate": 0.000474069767722015, + "loss": 1.9179, + "step": 1834 + }, + { + "epoch": 0.17919921875, + "grad_norm": 0.21640396118164062, + "learning_rate": 0.0004740366152735738, + "loss": 1.9204, + "step": 1835 + }, + { + "epoch": 0.179296875, + "grad_norm": 0.2559264302253723, + "learning_rate": 0.0004740034429431379, + "loss": 1.9145, + "step": 1836 + }, + { + "epoch": 0.17939453125, + "grad_norm": 0.24724526703357697, + "learning_rate": 0.00047397025073402116, + "loss": 1.9107, + "step": 1837 + }, + { + "epoch": 0.1794921875, + "grad_norm": 0.20995844900608063, + "learning_rate": 0.0004739370386495391, + "loss": 1.9408, + "step": 1838 + }, + { + "epoch": 0.17958984375, + "grad_norm": 0.24171245098114014, + "learning_rate": 0.00047390380669300923, + "loss": 1.9023, + "step": 1839 + }, + { + "epoch": 0.1796875, + "grad_norm": 0.2987540066242218, + "learning_rate": 0.00047387055486775123, + "loss": 1.932, + "step": 1840 + }, + { + "epoch": 0.17978515625, + "grad_norm": 0.2575046420097351, + "learning_rate": 0.0004738372831770866, + "loss": 1.8957, + "step": 1841 + }, + { + "epoch": 0.1798828125, + "grad_norm": 0.26643261313438416, + "learning_rate": 0.00047380399162433903, + "loss": 1.927, + "step": 1842 + }, + { + "epoch": 0.17998046875, + "grad_norm": 0.25104060769081116, + "learning_rate": 0.00047377068021283385, + "loss": 1.9379, + "step": 1843 + }, + { + "epoch": 0.180078125, + "grad_norm": 0.2125697284936905, + "learning_rate": 0.00047373734894589877, + "loss": 1.9262, + "step": 1844 + }, + { + "epoch": 0.18017578125, + "grad_norm": 0.3097136616706848, + "learning_rate": 0.0004737039978268631, + "loss": 1.9351, + "step": 1845 + }, + { + "epoch": 0.1802734375, + "grad_norm": 0.28434544801712036, + "learning_rate": 0.00047367062685905835, + "loss": 1.9399, + "step": 1846 + }, + { + "epoch": 0.18037109375, + "grad_norm": 0.2924177944660187, + "learning_rate": 0.00047363723604581815, + "loss": 1.8924, + "step": 1847 + }, + { + "epoch": 0.18046875, + "grad_norm": 0.2927818298339844, + "learning_rate": 0.0004736038253904778, + "loss": 1.8868, + "step": 1848 + }, + { + "epoch": 0.18056640625, + "grad_norm": 0.24946413934230804, + "learning_rate": 0.0004735703948963747, + "loss": 1.9145, + "step": 1849 + }, + { + "epoch": 0.1806640625, + "grad_norm": 0.2521260380744934, + "learning_rate": 0.00047353694456684844, + "loss": 1.9579, + "step": 1850 + }, + { + "epoch": 0.18076171875, + "grad_norm": 0.30722346901893616, + "learning_rate": 0.0004735034744052402, + "loss": 1.9279, + "step": 1851 + }, + { + "epoch": 0.180859375, + "grad_norm": 0.24488766491413116, + "learning_rate": 0.00047346998441489353, + "loss": 1.9055, + "step": 1852 + }, + { + "epoch": 0.18095703125, + "grad_norm": 0.29834532737731934, + "learning_rate": 0.00047343647459915364, + "loss": 1.9511, + "step": 1853 + }, + { + "epoch": 0.1810546875, + "grad_norm": 0.252108097076416, + "learning_rate": 0.000473402944961368, + "loss": 1.8903, + "step": 1854 + }, + { + "epoch": 0.18115234375, + "grad_norm": 0.21066221594810486, + "learning_rate": 0.00047336939550488575, + "loss": 1.9162, + "step": 1855 + }, + { + "epoch": 0.18125, + "grad_norm": 0.23784129321575165, + "learning_rate": 0.0004733358262330584, + "loss": 1.8824, + "step": 1856 + }, + { + "epoch": 0.18134765625, + "grad_norm": 0.20668673515319824, + "learning_rate": 0.000473302237149239, + "loss": 1.9197, + "step": 1857 + }, + { + "epoch": 0.1814453125, + "grad_norm": 0.21304477751255035, + "learning_rate": 0.00047326862825678296, + "loss": 1.883, + "step": 1858 + }, + { + "epoch": 0.18154296875, + "grad_norm": 0.240382581949234, + "learning_rate": 0.0004732349995590474, + "loss": 1.9065, + "step": 1859 + }, + { + "epoch": 0.181640625, + "grad_norm": 0.27892357110977173, + "learning_rate": 0.00047320135105939165, + "loss": 1.9338, + "step": 1860 + }, + { + "epoch": 0.18173828125, + "grad_norm": 0.22719697654247284, + "learning_rate": 0.0004731676827611768, + "loss": 1.8927, + "step": 1861 + }, + { + "epoch": 0.1818359375, + "grad_norm": 0.19273899495601654, + "learning_rate": 0.00047313399466776605, + "loss": 1.9132, + "step": 1862 + }, + { + "epoch": 0.18193359375, + "grad_norm": 0.2563495934009552, + "learning_rate": 0.0004731002867825246, + "loss": 1.9206, + "step": 1863 + }, + { + "epoch": 0.18203125, + "grad_norm": 0.29519474506378174, + "learning_rate": 0.0004730665591088195, + "loss": 1.9254, + "step": 1864 + }, + { + "epoch": 0.18212890625, + "grad_norm": 0.2755618095397949, + "learning_rate": 0.0004730328116500198, + "loss": 1.9348, + "step": 1865 + }, + { + "epoch": 0.1822265625, + "grad_norm": 0.25873810052871704, + "learning_rate": 0.00047299904440949663, + "loss": 1.9285, + "step": 1866 + }, + { + "epoch": 0.18232421875, + "grad_norm": 0.19655470550060272, + "learning_rate": 0.0004729652573906231, + "loss": 1.9268, + "step": 1867 + }, + { + "epoch": 0.182421875, + "grad_norm": 0.288585901260376, + "learning_rate": 0.00047293145059677403, + "loss": 1.9379, + "step": 1868 + }, + { + "epoch": 0.18251953125, + "grad_norm": 0.2560109794139862, + "learning_rate": 0.00047289762403132673, + "loss": 1.9258, + "step": 1869 + }, + { + "epoch": 0.1826171875, + "grad_norm": 0.2627304196357727, + "learning_rate": 0.00047286377769766, + "loss": 1.8885, + "step": 1870 + }, + { + "epoch": 0.18271484375, + "grad_norm": 0.3108537793159485, + "learning_rate": 0.0004728299115991547, + "loss": 1.9106, + "step": 1871 + }, + { + "epoch": 0.1828125, + "grad_norm": 0.24468936026096344, + "learning_rate": 0.00047279602573919386, + "loss": 1.9436, + "step": 1872 + }, + { + "epoch": 0.18291015625, + "grad_norm": 0.3040415048599243, + "learning_rate": 0.0004727621201211623, + "loss": 1.977, + "step": 1873 + }, + { + "epoch": 0.1830078125, + "grad_norm": 0.3214859366416931, + "learning_rate": 0.0004727281947484471, + "loss": 1.9259, + "step": 1874 + }, + { + "epoch": 0.18310546875, + "grad_norm": 0.3081786632537842, + "learning_rate": 0.0004726942496244369, + "loss": 1.8833, + "step": 1875 + }, + { + "epoch": 0.183203125, + "grad_norm": 0.2123297154903412, + "learning_rate": 0.0004726602847525225, + "loss": 1.9327, + "step": 1876 + }, + { + "epoch": 0.18330078125, + "grad_norm": 0.20726516842842102, + "learning_rate": 0.00047262630013609694, + "loss": 1.932, + "step": 1877 + }, + { + "epoch": 0.1833984375, + "grad_norm": 0.2831561863422394, + "learning_rate": 0.0004725922957785547, + "loss": 1.8966, + "step": 1878 + }, + { + "epoch": 0.18349609375, + "grad_norm": 0.25883564352989197, + "learning_rate": 0.00047255827168329255, + "loss": 1.8872, + "step": 1879 + }, + { + "epoch": 0.18359375, + "grad_norm": 0.24307870864868164, + "learning_rate": 0.00047252422785370936, + "loss": 1.935, + "step": 1880 + }, + { + "epoch": 0.18369140625, + "grad_norm": 0.2524338662624359, + "learning_rate": 0.00047249016429320567, + "loss": 1.887, + "step": 1881 + }, + { + "epoch": 0.1837890625, + "grad_norm": 0.22779789566993713, + "learning_rate": 0.0004724560810051843, + "loss": 1.9536, + "step": 1882 + }, + { + "epoch": 0.18388671875, + "grad_norm": 0.23726516962051392, + "learning_rate": 0.0004724219779930496, + "loss": 1.9267, + "step": 1883 + }, + { + "epoch": 0.183984375, + "grad_norm": 0.2780471742153168, + "learning_rate": 0.0004723878552602083, + "loss": 1.9097, + "step": 1884 + }, + { + "epoch": 0.18408203125, + "grad_norm": 0.2215142548084259, + "learning_rate": 0.000472353712810069, + "loss": 1.8851, + "step": 1885 + }, + { + "epoch": 0.1841796875, + "grad_norm": 0.22764411568641663, + "learning_rate": 0.0004723195506460422, + "loss": 1.9108, + "step": 1886 + }, + { + "epoch": 0.18427734375, + "grad_norm": 0.2349424958229065, + "learning_rate": 0.0004722853687715404, + "loss": 1.9079, + "step": 1887 + }, + { + "epoch": 0.184375, + "grad_norm": 0.23532803356647491, + "learning_rate": 0.00047225116718997804, + "loss": 1.9198, + "step": 1888 + }, + { + "epoch": 0.18447265625, + "grad_norm": 0.2555566728115082, + "learning_rate": 0.00047221694590477147, + "loss": 1.8961, + "step": 1889 + }, + { + "epoch": 0.1845703125, + "grad_norm": 0.30815714597702026, + "learning_rate": 0.00047218270491933934, + "loss": 1.8792, + "step": 1890 + }, + { + "epoch": 0.18466796875, + "grad_norm": 0.21387292444705963, + "learning_rate": 0.00047214844423710173, + "loss": 1.918, + "step": 1891 + }, + { + "epoch": 0.184765625, + "grad_norm": 0.24553145468235016, + "learning_rate": 0.00047211416386148123, + "loss": 1.8908, + "step": 1892 + }, + { + "epoch": 0.18486328125, + "grad_norm": 0.3307762145996094, + "learning_rate": 0.00047207986379590197, + "loss": 1.9298, + "step": 1893 + }, + { + "epoch": 0.1849609375, + "grad_norm": 0.22250519692897797, + "learning_rate": 0.00047204554404379036, + "loss": 1.886, + "step": 1894 + }, + { + "epoch": 0.18505859375, + "grad_norm": 0.2863367199897766, + "learning_rate": 0.0004720112046085745, + "loss": 1.8674, + "step": 1895 + }, + { + "epoch": 0.18515625, + "grad_norm": 0.34826767444610596, + "learning_rate": 0.00047197684549368465, + "loss": 1.9282, + "step": 1896 + }, + { + "epoch": 0.18525390625, + "grad_norm": 0.289986252784729, + "learning_rate": 0.00047194246670255295, + "loss": 1.9078, + "step": 1897 + }, + { + "epoch": 0.1853515625, + "grad_norm": 0.29067522287368774, + "learning_rate": 0.0004719080682386137, + "loss": 1.8943, + "step": 1898 + }, + { + "epoch": 0.18544921875, + "grad_norm": 0.2805711627006531, + "learning_rate": 0.0004718736501053028, + "loss": 1.8729, + "step": 1899 + }, + { + "epoch": 0.185546875, + "grad_norm": 0.1988450288772583, + "learning_rate": 0.0004718392123060584, + "loss": 1.897, + "step": 1900 + }, + { + "epoch": 0.18564453125, + "grad_norm": 0.28334251046180725, + "learning_rate": 0.0004718047548443206, + "loss": 1.8796, + "step": 1901 + }, + { + "epoch": 0.1857421875, + "grad_norm": 0.24370847642421722, + "learning_rate": 0.00047177027772353134, + "loss": 1.9136, + "step": 1902 + }, + { + "epoch": 0.18583984375, + "grad_norm": 0.25666430592536926, + "learning_rate": 0.0004717357809471344, + "loss": 1.9396, + "step": 1903 + }, + { + "epoch": 0.1859375, + "grad_norm": 0.3168308138847351, + "learning_rate": 0.00047170126451857603, + "loss": 1.8627, + "step": 1904 + }, + { + "epoch": 0.18603515625, + "grad_norm": 0.3242661952972412, + "learning_rate": 0.0004716667284413039, + "loss": 1.9534, + "step": 1905 + }, + { + "epoch": 0.1861328125, + "grad_norm": 0.23754087090492249, + "learning_rate": 0.0004716321727187678, + "loss": 1.935, + "step": 1906 + }, + { + "epoch": 0.18623046875, + "grad_norm": 0.2671366035938263, + "learning_rate": 0.0004715975973544198, + "loss": 1.9543, + "step": 1907 + }, + { + "epoch": 0.186328125, + "grad_norm": 0.2637098729610443, + "learning_rate": 0.00047156300235171353, + "loss": 1.8575, + "step": 1908 + }, + { + "epoch": 0.18642578125, + "grad_norm": 0.22229737043380737, + "learning_rate": 0.0004715283877141046, + "loss": 1.8757, + "step": 1909 + }, + { + "epoch": 0.1865234375, + "grad_norm": 0.2838973104953766, + "learning_rate": 0.00047149375344505084, + "loss": 1.9114, + "step": 1910 + }, + { + "epoch": 0.18662109375, + "grad_norm": 0.30752044916152954, + "learning_rate": 0.00047145909954801195, + "loss": 1.8879, + "step": 1911 + }, + { + "epoch": 0.18671875, + "grad_norm": 0.2020583301782608, + "learning_rate": 0.0004714244260264495, + "loss": 1.8957, + "step": 1912 + }, + { + "epoch": 0.18681640625, + "grad_norm": 0.25839710235595703, + "learning_rate": 0.000471389732883827, + "loss": 1.9356, + "step": 1913 + }, + { + "epoch": 0.1869140625, + "grad_norm": 0.2770155668258667, + "learning_rate": 0.0004713550201236101, + "loss": 1.9127, + "step": 1914 + }, + { + "epoch": 0.18701171875, + "grad_norm": 0.20789627730846405, + "learning_rate": 0.0004713202877492661, + "loss": 1.9254, + "step": 1915 + }, + { + "epoch": 0.187109375, + "grad_norm": 0.2471790611743927, + "learning_rate": 0.00047128553576426477, + "loss": 1.958, + "step": 1916 + }, + { + "epoch": 0.18720703125, + "grad_norm": 0.2624439597129822, + "learning_rate": 0.0004712507641720772, + "loss": 1.9403, + "step": 1917 + }, + { + "epoch": 0.1873046875, + "grad_norm": 0.2492697685956955, + "learning_rate": 0.00047121597297617704, + "loss": 1.8928, + "step": 1918 + }, + { + "epoch": 0.18740234375, + "grad_norm": 0.20261016488075256, + "learning_rate": 0.0004711811621800394, + "loss": 1.925, + "step": 1919 + }, + { + "epoch": 0.1875, + "grad_norm": 0.20547007024288177, + "learning_rate": 0.0004711463317871417, + "loss": 1.9058, + "step": 1920 + }, + { + "epoch": 0.18759765625, + "grad_norm": 0.24086418747901917, + "learning_rate": 0.0004711114818009632, + "loss": 1.9377, + "step": 1921 + }, + { + "epoch": 0.1876953125, + "grad_norm": 0.28701573610305786, + "learning_rate": 0.00047107661222498497, + "loss": 1.8958, + "step": 1922 + }, + { + "epoch": 0.18779296875, + "grad_norm": 0.2246810346841812, + "learning_rate": 0.0004710417230626904, + "loss": 1.8759, + "step": 1923 + }, + { + "epoch": 0.187890625, + "grad_norm": 0.1889760047197342, + "learning_rate": 0.00047100681431756433, + "loss": 1.947, + "step": 1924 + }, + { + "epoch": 0.18798828125, + "grad_norm": 0.22808609902858734, + "learning_rate": 0.0004709718859930941, + "loss": 1.9428, + "step": 1925 + }, + { + "epoch": 0.1880859375, + "grad_norm": 0.2610793709754944, + "learning_rate": 0.00047093693809276843, + "loss": 1.964, + "step": 1926 + }, + { + "epoch": 0.18818359375, + "grad_norm": 0.30671024322509766, + "learning_rate": 0.00047090197062007864, + "loss": 1.9101, + "step": 1927 + }, + { + "epoch": 0.18828125, + "grad_norm": 0.27946990728378296, + "learning_rate": 0.0004708669835785175, + "loss": 1.9487, + "step": 1928 + }, + { + "epoch": 0.18837890625, + "grad_norm": 0.2439918965101242, + "learning_rate": 0.0004708319769715799, + "loss": 1.9037, + "step": 1929 + }, + { + "epoch": 0.1884765625, + "grad_norm": 0.2517712414264679, + "learning_rate": 0.0004707969508027627, + "loss": 1.9059, + "step": 1930 + }, + { + "epoch": 0.18857421875, + "grad_norm": 0.2609822750091553, + "learning_rate": 0.0004707619050755648, + "loss": 1.9119, + "step": 1931 + }, + { + "epoch": 0.188671875, + "grad_norm": 0.2705680727958679, + "learning_rate": 0.0004707268397934868, + "loss": 1.9264, + "step": 1932 + }, + { + "epoch": 0.18876953125, + "grad_norm": 0.27703019976615906, + "learning_rate": 0.00047069175496003147, + "loss": 1.8859, + "step": 1933 + }, + { + "epoch": 0.1888671875, + "grad_norm": 0.4434245228767395, + "learning_rate": 0.00047065665057870355, + "loss": 2.0193, + "step": 1934 + }, + { + "epoch": 0.18896484375, + "grad_norm": 0.25824934244155884, + "learning_rate": 0.0004706215266530096, + "loss": 1.9491, + "step": 1935 + }, + { + "epoch": 0.1890625, + "grad_norm": 0.3010888397693634, + "learning_rate": 0.00047058638318645815, + "loss": 1.9461, + "step": 1936 + }, + { + "epoch": 0.18916015625, + "grad_norm": 0.2592420279979706, + "learning_rate": 0.0004705512201825597, + "loss": 1.9329, + "step": 1937 + }, + { + "epoch": 0.1892578125, + "grad_norm": 0.2857649326324463, + "learning_rate": 0.0004705160376448269, + "loss": 1.913, + "step": 1938 + }, + { + "epoch": 0.18935546875, + "grad_norm": 0.3039281964302063, + "learning_rate": 0.000470480835576774, + "loss": 1.9253, + "step": 1939 + }, + { + "epoch": 0.189453125, + "grad_norm": 0.23744364082813263, + "learning_rate": 0.00047044561398191744, + "loss": 1.9015, + "step": 1940 + }, + { + "epoch": 0.18955078125, + "grad_norm": 0.24613125622272491, + "learning_rate": 0.0004704103728637756, + "loss": 1.9214, + "step": 1941 + }, + { + "epoch": 0.1896484375, + "grad_norm": 0.2505404055118561, + "learning_rate": 0.0004703751122258686, + "loss": 1.9532, + "step": 1942 + }, + { + "epoch": 0.18974609375, + "grad_norm": 0.2505582571029663, + "learning_rate": 0.0004703398320717188, + "loss": 1.932, + "step": 1943 + }, + { + "epoch": 0.18984375, + "grad_norm": 0.2656853199005127, + "learning_rate": 0.00047030453240485035, + "loss": 1.8801, + "step": 1944 + }, + { + "epoch": 0.18994140625, + "grad_norm": 0.2200378179550171, + "learning_rate": 0.0004702692132287892, + "loss": 1.9601, + "step": 1945 + }, + { + "epoch": 0.1900390625, + "grad_norm": 0.260431170463562, + "learning_rate": 0.00047023387454706375, + "loss": 1.9112, + "step": 1946 + }, + { + "epoch": 0.19013671875, + "grad_norm": 0.2725105881690979, + "learning_rate": 0.00047019851636320384, + "loss": 1.9616, + "step": 1947 + }, + { + "epoch": 0.190234375, + "grad_norm": 0.21269826591014862, + "learning_rate": 0.0004701631386807414, + "loss": 1.9063, + "step": 1948 + }, + { + "epoch": 0.19033203125, + "grad_norm": 0.2665625512599945, + "learning_rate": 0.0004701277415032104, + "loss": 1.9361, + "step": 1949 + }, + { + "epoch": 0.1904296875, + "grad_norm": 0.2765948176383972, + "learning_rate": 0.0004700923248341467, + "loss": 1.9528, + "step": 1950 + }, + { + "epoch": 0.19052734375, + "grad_norm": 0.22930294275283813, + "learning_rate": 0.00047005688867708814, + "loss": 1.8957, + "step": 1951 + }, + { + "epoch": 0.190625, + "grad_norm": 0.22256456315517426, + "learning_rate": 0.00047002143303557435, + "loss": 1.9489, + "step": 1952 + }, + { + "epoch": 0.19072265625, + "grad_norm": 0.24130624532699585, + "learning_rate": 0.00046998595791314726, + "loss": 1.9574, + "step": 1953 + }, + { + "epoch": 0.1908203125, + "grad_norm": 0.292375773191452, + "learning_rate": 0.00046995046331335026, + "loss": 1.9176, + "step": 1954 + }, + { + "epoch": 0.19091796875, + "grad_norm": 0.3528386950492859, + "learning_rate": 0.0004699149492397292, + "loss": 1.9511, + "step": 1955 + }, + { + "epoch": 0.191015625, + "grad_norm": 0.26164865493774414, + "learning_rate": 0.00046987941569583147, + "loss": 1.9094, + "step": 1956 + }, + { + "epoch": 0.19111328125, + "grad_norm": 0.27652114629745483, + "learning_rate": 0.0004698438626852066, + "loss": 1.9011, + "step": 1957 + }, + { + "epoch": 0.1912109375, + "grad_norm": 0.31081855297088623, + "learning_rate": 0.00046980829021140596, + "loss": 1.9041, + "step": 1958 + }, + { + "epoch": 0.19130859375, + "grad_norm": 0.29934486746788025, + "learning_rate": 0.00046977269827798304, + "loss": 1.9301, + "step": 1959 + }, + { + "epoch": 0.19140625, + "grad_norm": 0.31369319558143616, + "learning_rate": 0.000469737086888493, + "loss": 1.9194, + "step": 1960 + }, + { + "epoch": 0.19150390625, + "grad_norm": 0.21880319714546204, + "learning_rate": 0.00046970145604649326, + "loss": 1.9029, + "step": 1961 + }, + { + "epoch": 0.1916015625, + "grad_norm": 0.21874625980854034, + "learning_rate": 0.00046966580575554304, + "loss": 1.9073, + "step": 1962 + }, + { + "epoch": 0.19169921875, + "grad_norm": 0.27970215678215027, + "learning_rate": 0.0004696301360192033, + "loss": 1.9589, + "step": 1963 + }, + { + "epoch": 0.191796875, + "grad_norm": 0.22823897004127502, + "learning_rate": 0.00046959444684103725, + "loss": 1.8801, + "step": 1964 + }, + { + "epoch": 0.19189453125, + "grad_norm": 0.22956214845180511, + "learning_rate": 0.0004695587382246099, + "loss": 1.908, + "step": 1965 + }, + { + "epoch": 0.1919921875, + "grad_norm": 0.2367463856935501, + "learning_rate": 0.00046952301017348826, + "loss": 1.9049, + "step": 1966 + }, + { + "epoch": 0.19208984375, + "grad_norm": 0.2357739508152008, + "learning_rate": 0.0004694872626912412, + "loss": 1.9173, + "step": 1967 + }, + { + "epoch": 0.1921875, + "grad_norm": 0.25406715273857117, + "learning_rate": 0.0004694514957814396, + "loss": 1.9309, + "step": 1968 + }, + { + "epoch": 0.19228515625, + "grad_norm": 0.2415982335805893, + "learning_rate": 0.0004694157094476563, + "loss": 1.8976, + "step": 1969 + }, + { + "epoch": 0.1923828125, + "grad_norm": 0.2648661434650421, + "learning_rate": 0.000469379903693466, + "loss": 1.9001, + "step": 1970 + }, + { + "epoch": 0.19248046875, + "grad_norm": 0.186269611120224, + "learning_rate": 0.00046934407852244534, + "loss": 1.8917, + "step": 1971 + }, + { + "epoch": 0.192578125, + "grad_norm": 0.30947110056877136, + "learning_rate": 0.00046930823393817296, + "loss": 1.8892, + "step": 1972 + }, + { + "epoch": 0.19267578125, + "grad_norm": 0.2893858850002289, + "learning_rate": 0.00046927236994422945, + "loss": 1.9225, + "step": 1973 + }, + { + "epoch": 0.1927734375, + "grad_norm": 0.24984197318553925, + "learning_rate": 0.0004692364865441973, + "loss": 1.9367, + "step": 1974 + }, + { + "epoch": 0.19287109375, + "grad_norm": 0.27524879574775696, + "learning_rate": 0.0004692005837416609, + "loss": 1.9141, + "step": 1975 + }, + { + "epoch": 0.19296875, + "grad_norm": 0.22581790387630463, + "learning_rate": 0.00046916466154020656, + "loss": 1.9093, + "step": 1976 + }, + { + "epoch": 0.19306640625, + "grad_norm": 0.2497783899307251, + "learning_rate": 0.0004691287199434228, + "loss": 1.9136, + "step": 1977 + }, + { + "epoch": 0.1931640625, + "grad_norm": 0.22558996081352234, + "learning_rate": 0.0004690927589548997, + "loss": 1.9372, + "step": 1978 + }, + { + "epoch": 0.19326171875, + "grad_norm": 0.2091677039861679, + "learning_rate": 0.0004690567785782295, + "loss": 1.9425, + "step": 1979 + }, + { + "epoch": 0.193359375, + "grad_norm": 0.26007091999053955, + "learning_rate": 0.0004690207788170063, + "loss": 1.8893, + "step": 1980 + }, + { + "epoch": 0.19345703125, + "grad_norm": 0.2551250755786896, + "learning_rate": 0.0004689847596748261, + "loss": 1.924, + "step": 1981 + }, + { + "epoch": 0.1935546875, + "grad_norm": 0.30021947622299194, + "learning_rate": 0.000468948721155287, + "loss": 1.9103, + "step": 1982 + }, + { + "epoch": 0.19365234375, + "grad_norm": 0.2922191917896271, + "learning_rate": 0.0004689126632619889, + "loss": 1.9284, + "step": 1983 + }, + { + "epoch": 0.19375, + "grad_norm": 0.24520905315876007, + "learning_rate": 0.00046887658599853373, + "loss": 1.8806, + "step": 1984 + }, + { + "epoch": 0.19384765625, + "grad_norm": 0.27852240204811096, + "learning_rate": 0.00046884048936852513, + "loss": 1.9375, + "step": 1985 + }, + { + "epoch": 0.1939453125, + "grad_norm": 0.26879197359085083, + "learning_rate": 0.0004688043733755689, + "loss": 1.9054, + "step": 1986 + }, + { + "epoch": 0.19404296875, + "grad_norm": 0.2576071321964264, + "learning_rate": 0.00046876823802327274, + "loss": 1.9758, + "step": 1987 + }, + { + "epoch": 0.194140625, + "grad_norm": 0.24778784811496735, + "learning_rate": 0.0004687320833152463, + "loss": 1.8767, + "step": 1988 + }, + { + "epoch": 0.19423828125, + "grad_norm": 0.2549983859062195, + "learning_rate": 0.000468695909255101, + "loss": 1.9532, + "step": 1989 + }, + { + "epoch": 0.1943359375, + "grad_norm": 0.24673303961753845, + "learning_rate": 0.00046865971584645027, + "loss": 1.9304, + "step": 1990 + }, + { + "epoch": 0.19443359375, + "grad_norm": 0.23759478330612183, + "learning_rate": 0.00046862350309290977, + "loss": 1.8887, + "step": 1991 + }, + { + "epoch": 0.19453125, + "grad_norm": 0.28576332330703735, + "learning_rate": 0.00046858727099809643, + "loss": 1.8766, + "step": 1992 + }, + { + "epoch": 0.19462890625, + "grad_norm": 0.23105448484420776, + "learning_rate": 0.00046855101956563, + "loss": 1.8956, + "step": 1993 + }, + { + "epoch": 0.1947265625, + "grad_norm": 0.27505502104759216, + "learning_rate": 0.0004685147487991312, + "loss": 1.8915, + "step": 1994 + }, + { + "epoch": 0.19482421875, + "grad_norm": 0.2024049013853073, + "learning_rate": 0.00046847845870222354, + "loss": 1.8735, + "step": 1995 + }, + { + "epoch": 0.194921875, + "grad_norm": 0.26097309589385986, + "learning_rate": 0.0004684421492785318, + "loss": 1.9298, + "step": 1996 + }, + { + "epoch": 0.19501953125, + "grad_norm": 0.2358173280954361, + "learning_rate": 0.0004684058205316832, + "loss": 1.9121, + "step": 1997 + }, + { + "epoch": 0.1951171875, + "grad_norm": 0.2725095748901367, + "learning_rate": 0.00046836947246530646, + "loss": 1.8684, + "step": 1998 + }, + { + "epoch": 0.19521484375, + "grad_norm": 0.25193265080451965, + "learning_rate": 0.0004683331050830326, + "loss": 1.9188, + "step": 1999 + }, + { + "epoch": 0.1953125, + "grad_norm": 0.23056617379188538, + "learning_rate": 0.00046829671838849424, + "loss": 1.8628, + "step": 2000 + }, + { + "epoch": 0.19541015625, + "grad_norm": 0.21980516612529755, + "learning_rate": 0.00046826031238532623, + "loss": 1.9392, + "step": 2001 + }, + { + "epoch": 0.1955078125, + "grad_norm": 0.244726300239563, + "learning_rate": 0.0004682238870771651, + "loss": 1.9079, + "step": 2002 + }, + { + "epoch": 0.19560546875, + "grad_norm": 0.21440590918064117, + "learning_rate": 0.00046818744246764955, + "loss": 1.9059, + "step": 2003 + }, + { + "epoch": 0.195703125, + "grad_norm": 0.25402185320854187, + "learning_rate": 0.0004681509785604199, + "loss": 1.9226, + "step": 2004 + }, + { + "epoch": 0.19580078125, + "grad_norm": 0.24775418639183044, + "learning_rate": 0.00046811449535911863, + "loss": 1.8361, + "step": 2005 + }, + { + "epoch": 0.1958984375, + "grad_norm": 0.2603761553764343, + "learning_rate": 0.0004680779928673902, + "loss": 1.9252, + "step": 2006 + }, + { + "epoch": 0.19599609375, + "grad_norm": 0.25276273488998413, + "learning_rate": 0.0004680414710888808, + "loss": 1.9072, + "step": 2007 + }, + { + "epoch": 0.19609375, + "grad_norm": 0.23971259593963623, + "learning_rate": 0.00046800493002723854, + "loss": 1.9073, + "step": 2008 + }, + { + "epoch": 0.19619140625, + "grad_norm": 0.21876166760921478, + "learning_rate": 0.00046796836968611373, + "loss": 1.8839, + "step": 2009 + }, + { + "epoch": 0.1962890625, + "grad_norm": 0.22767984867095947, + "learning_rate": 0.00046793179006915825, + "loss": 1.8852, + "step": 2010 + }, + { + "epoch": 0.19638671875, + "grad_norm": 0.1997627317905426, + "learning_rate": 0.0004678951911800262, + "loss": 1.8831, + "step": 2011 + }, + { + "epoch": 0.196484375, + "grad_norm": 0.23563456535339355, + "learning_rate": 0.00046785857302237345, + "loss": 1.8791, + "step": 2012 + }, + { + "epoch": 0.19658203125, + "grad_norm": 0.1879824995994568, + "learning_rate": 0.0004678219355998578, + "loss": 1.8986, + "step": 2013 + }, + { + "epoch": 0.1966796875, + "grad_norm": 0.26914268732070923, + "learning_rate": 0.00046778527891613907, + "loss": 1.9269, + "step": 2014 + }, + { + "epoch": 0.19677734375, + "grad_norm": 0.20528461039066315, + "learning_rate": 0.000467748602974879, + "loss": 1.9475, + "step": 2015 + }, + { + "epoch": 0.196875, + "grad_norm": 0.260937362909317, + "learning_rate": 0.0004677119077797409, + "loss": 1.9036, + "step": 2016 + }, + { + "epoch": 0.19697265625, + "grad_norm": 0.233351469039917, + "learning_rate": 0.0004676751933343906, + "loss": 1.9155, + "step": 2017 + }, + { + "epoch": 0.1970703125, + "grad_norm": 0.24918298423290253, + "learning_rate": 0.0004676384596424954, + "loss": 1.9169, + "step": 2018 + }, + { + "epoch": 0.19716796875, + "grad_norm": 0.331938236951828, + "learning_rate": 0.00046760170670772473, + "loss": 1.8655, + "step": 2019 + }, + { + "epoch": 0.197265625, + "grad_norm": 0.31642693281173706, + "learning_rate": 0.0004675649345337498, + "loss": 1.926, + "step": 2020 + }, + { + "epoch": 0.19736328125, + "grad_norm": 0.28673624992370605, + "learning_rate": 0.00046752814312424394, + "loss": 1.8992, + "step": 2021 + }, + { + "epoch": 0.1974609375, + "grad_norm": 0.27498120069503784, + "learning_rate": 0.0004674913324828822, + "loss": 1.9125, + "step": 2022 + }, + { + "epoch": 0.19755859375, + "grad_norm": 0.3634410500526428, + "learning_rate": 0.0004674545026133417, + "loss": 1.9216, + "step": 2023 + }, + { + "epoch": 0.19765625, + "grad_norm": 0.2195715606212616, + "learning_rate": 0.00046741765351930137, + "loss": 1.9071, + "step": 2024 + }, + { + "epoch": 0.19775390625, + "grad_norm": 0.30638429522514343, + "learning_rate": 0.0004673807852044421, + "loss": 1.9037, + "step": 2025 + }, + { + "epoch": 0.1978515625, + "grad_norm": 0.30112913250923157, + "learning_rate": 0.0004673438976724468, + "loss": 1.9117, + "step": 2026 + }, + { + "epoch": 0.19794921875, + "grad_norm": 0.2405427247285843, + "learning_rate": 0.00046730699092700003, + "loss": 1.8978, + "step": 2027 + }, + { + "epoch": 0.198046875, + "grad_norm": 0.38917067646980286, + "learning_rate": 0.0004672700649717886, + "loss": 1.9055, + "step": 2028 + }, + { + "epoch": 0.19814453125, + "grad_norm": 0.31037402153015137, + "learning_rate": 0.000467233119810501, + "loss": 1.8982, + "step": 2029 + }, + { + "epoch": 0.1982421875, + "grad_norm": 0.31107863783836365, + "learning_rate": 0.00046719615544682784, + "loss": 1.8812, + "step": 2030 + }, + { + "epoch": 0.19833984375, + "grad_norm": 0.3467962145805359, + "learning_rate": 0.00046715917188446134, + "loss": 1.8957, + "step": 2031 + }, + { + "epoch": 0.1984375, + "grad_norm": 0.25533223152160645, + "learning_rate": 0.0004671221691270961, + "loss": 1.9409, + "step": 2032 + }, + { + "epoch": 0.19853515625, + "grad_norm": 0.3249166011810303, + "learning_rate": 0.0004670851471784281, + "loss": 1.9037, + "step": 2033 + }, + { + "epoch": 0.1986328125, + "grad_norm": 0.2961365282535553, + "learning_rate": 0.00046704810604215565, + "loss": 1.8862, + "step": 2034 + }, + { + "epoch": 0.19873046875, + "grad_norm": 0.30742213129997253, + "learning_rate": 0.0004670110457219788, + "loss": 1.8787, + "step": 2035 + }, + { + "epoch": 0.198828125, + "grad_norm": 0.20632188022136688, + "learning_rate": 0.00046697396622159956, + "loss": 1.8716, + "step": 2036 + }, + { + "epoch": 0.19892578125, + "grad_norm": 0.3098668158054352, + "learning_rate": 0.0004669368675447218, + "loss": 1.9131, + "step": 2037 + }, + { + "epoch": 0.1990234375, + "grad_norm": 0.210395947098732, + "learning_rate": 0.00046689974969505143, + "loss": 1.8658, + "step": 2038 + }, + { + "epoch": 0.19912109375, + "grad_norm": 0.2865216135978699, + "learning_rate": 0.00046686261267629605, + "loss": 1.9041, + "step": 2039 + }, + { + "epoch": 0.19921875, + "grad_norm": 0.2984568774700165, + "learning_rate": 0.00046682545649216544, + "loss": 1.9122, + "step": 2040 + }, + { + "epoch": 0.19931640625, + "grad_norm": 0.285224586725235, + "learning_rate": 0.00046678828114637126, + "loss": 1.9139, + "step": 2041 + }, + { + "epoch": 0.1994140625, + "grad_norm": 0.23279589414596558, + "learning_rate": 0.0004667510866426268, + "loss": 1.8941, + "step": 2042 + }, + { + "epoch": 0.19951171875, + "grad_norm": 0.20549869537353516, + "learning_rate": 0.0004667138729846475, + "loss": 1.8947, + "step": 2043 + }, + { + "epoch": 0.199609375, + "grad_norm": 0.26036345958709717, + "learning_rate": 0.0004666766401761508, + "loss": 1.9111, + "step": 2044 + }, + { + "epoch": 0.19970703125, + "grad_norm": 0.2697281837463379, + "learning_rate": 0.00046663938822085583, + "loss": 1.909, + "step": 2045 + }, + { + "epoch": 0.1998046875, + "grad_norm": 0.23096928000450134, + "learning_rate": 0.00046660211712248374, + "loss": 1.9102, + "step": 2046 + }, + { + "epoch": 0.19990234375, + "grad_norm": 0.21505551040172577, + "learning_rate": 0.00046656482688475754, + "loss": 1.9257, + "step": 2047 + }, + { + "epoch": 0.2, + "grad_norm": 0.2264963835477829, + "learning_rate": 0.0004665275175114024, + "loss": 1.9256, + "step": 2048 + }, + { + "epoch": 0.20009765625, + "grad_norm": 0.29406315088272095, + "learning_rate": 0.0004664901890061449, + "loss": 1.9512, + "step": 2049 + }, + { + "epoch": 0.2001953125, + "grad_norm": 0.2077530175447464, + "learning_rate": 0.0004664528413727139, + "loss": 1.9162, + "step": 2050 + }, + { + "epoch": 0.20029296875, + "grad_norm": 0.23434431850910187, + "learning_rate": 0.00046641547461484035, + "loss": 1.8639, + "step": 2051 + }, + { + "epoch": 0.200390625, + "grad_norm": 0.3102989196777344, + "learning_rate": 0.0004663780887362566, + "loss": 1.8729, + "step": 2052 + }, + { + "epoch": 0.20048828125, + "grad_norm": 0.2750473618507385, + "learning_rate": 0.0004663406837406973, + "loss": 1.8936, + "step": 2053 + }, + { + "epoch": 0.2005859375, + "grad_norm": 0.24825125932693481, + "learning_rate": 0.00046630325963189866, + "loss": 1.8843, + "step": 2054 + }, + { + "epoch": 0.20068359375, + "grad_norm": 0.2034183144569397, + "learning_rate": 0.0004662658164135993, + "loss": 1.9049, + "step": 2055 + }, + { + "epoch": 0.20078125, + "grad_norm": 0.26106199622154236, + "learning_rate": 0.0004662283540895394, + "loss": 1.9034, + "step": 2056 + }, + { + "epoch": 0.20087890625, + "grad_norm": 0.22128944098949432, + "learning_rate": 0.0004661908726634609, + "loss": 1.8846, + "step": 2057 + }, + { + "epoch": 0.2009765625, + "grad_norm": 0.19975769519805908, + "learning_rate": 0.0004661533721391081, + "loss": 1.9336, + "step": 2058 + }, + { + "epoch": 0.20107421875, + "grad_norm": 0.21513822674751282, + "learning_rate": 0.0004661158525202269, + "loss": 1.8887, + "step": 2059 + }, + { + "epoch": 0.201171875, + "grad_norm": 0.2806141972541809, + "learning_rate": 0.00046607831381056513, + "loss": 1.8993, + "step": 2060 + }, + { + "epoch": 0.20126953125, + "grad_norm": 0.24645020067691803, + "learning_rate": 0.00046604075601387263, + "loss": 1.9206, + "step": 2061 + }, + { + "epoch": 0.2013671875, + "grad_norm": 0.19376933574676514, + "learning_rate": 0.0004660031791339011, + "loss": 1.9024, + "step": 2062 + }, + { + "epoch": 0.20146484375, + "grad_norm": 0.3149716556072235, + "learning_rate": 0.00046596558317440415, + "loss": 1.9294, + "step": 2063 + }, + { + "epoch": 0.2015625, + "grad_norm": 0.2759716808795929, + "learning_rate": 0.00046592796813913715, + "loss": 1.8748, + "step": 2064 + }, + { + "epoch": 0.20166015625, + "grad_norm": 0.2037348598241806, + "learning_rate": 0.0004658903340318577, + "loss": 1.8803, + "step": 2065 + }, + { + "epoch": 0.2017578125, + "grad_norm": 0.21185782551765442, + "learning_rate": 0.0004658526808563248, + "loss": 1.9478, + "step": 2066 + }, + { + "epoch": 0.20185546875, + "grad_norm": 0.2070467174053192, + "learning_rate": 0.0004658150086163001, + "loss": 1.9131, + "step": 2067 + }, + { + "epoch": 0.201953125, + "grad_norm": 0.22816209495067596, + "learning_rate": 0.00046577731731554646, + "loss": 1.9053, + "step": 2068 + }, + { + "epoch": 0.20205078125, + "grad_norm": 0.2772216796875, + "learning_rate": 0.0004657396069578289, + "loss": 1.9048, + "step": 2069 + }, + { + "epoch": 0.2021484375, + "grad_norm": 0.27881917357444763, + "learning_rate": 0.0004657018775469145, + "loss": 1.9161, + "step": 2070 + }, + { + "epoch": 0.20224609375, + "grad_norm": 0.2583853602409363, + "learning_rate": 0.0004656641290865719, + "loss": 1.8961, + "step": 2071 + }, + { + "epoch": 0.20234375, + "grad_norm": 0.2798794209957123, + "learning_rate": 0.00046562636158057203, + "loss": 1.8912, + "step": 2072 + }, + { + "epoch": 0.20244140625, + "grad_norm": 0.25222232937812805, + "learning_rate": 0.0004655885750326874, + "loss": 1.9522, + "step": 2073 + }, + { + "epoch": 0.2025390625, + "grad_norm": 0.2819882929325104, + "learning_rate": 0.0004655507694466925, + "loss": 1.8891, + "step": 2074 + }, + { + "epoch": 0.20263671875, + "grad_norm": 0.21219240128993988, + "learning_rate": 0.0004655129448263639, + "loss": 1.9383, + "step": 2075 + }, + { + "epoch": 0.202734375, + "grad_norm": 0.2394660860300064, + "learning_rate": 0.00046547510117547997, + "loss": 1.9034, + "step": 2076 + }, + { + "epoch": 0.20283203125, + "grad_norm": 0.22925956547260284, + "learning_rate": 0.00046543723849782085, + "loss": 1.9189, + "step": 2077 + }, + { + "epoch": 0.2029296875, + "grad_norm": 0.27168065309524536, + "learning_rate": 0.00046539935679716876, + "loss": 1.8803, + "step": 2078 + }, + { + "epoch": 0.20302734375, + "grad_norm": 0.3125360906124115, + "learning_rate": 0.00046536145607730767, + "loss": 1.9144, + "step": 2079 + }, + { + "epoch": 0.203125, + "grad_norm": 0.2742547392845154, + "learning_rate": 0.00046532353634202355, + "loss": 1.9011, + "step": 2080 + }, + { + "epoch": 0.20322265625, + "grad_norm": 0.30199679732322693, + "learning_rate": 0.0004652855975951044, + "loss": 1.8952, + "step": 2081 + }, + { + "epoch": 0.2033203125, + "grad_norm": 0.2881486117839813, + "learning_rate": 0.0004652476398403397, + "loss": 1.8888, + "step": 2082 + }, + { + "epoch": 0.20341796875, + "grad_norm": 0.30965685844421387, + "learning_rate": 0.0004652096630815212, + "loss": 1.9286, + "step": 2083 + }, + { + "epoch": 0.203515625, + "grad_norm": 0.38387784361839294, + "learning_rate": 0.00046517166732244254, + "loss": 1.922, + "step": 2084 + }, + { + "epoch": 0.20361328125, + "grad_norm": 0.35733339190483093, + "learning_rate": 0.00046513365256689907, + "loss": 1.9086, + "step": 2085 + }, + { + "epoch": 0.2037109375, + "grad_norm": 0.28068143129348755, + "learning_rate": 0.00046509561881868814, + "loss": 1.8819, + "step": 2086 + }, + { + "epoch": 0.20380859375, + "grad_norm": 0.2930566668510437, + "learning_rate": 0.00046505756608160895, + "loss": 1.8737, + "step": 2087 + }, + { + "epoch": 0.20390625, + "grad_norm": 0.2810053527355194, + "learning_rate": 0.0004650194943594627, + "loss": 1.9112, + "step": 2088 + }, + { + "epoch": 0.20400390625, + "grad_norm": 0.2667030394077301, + "learning_rate": 0.00046498140365605243, + "loss": 1.9051, + "step": 2089 + }, + { + "epoch": 0.2041015625, + "grad_norm": 0.20515574514865875, + "learning_rate": 0.00046494329397518284, + "loss": 1.9115, + "step": 2090 + }, + { + "epoch": 0.20419921875, + "grad_norm": 0.24670284986495972, + "learning_rate": 0.00046490516532066107, + "loss": 1.885, + "step": 2091 + }, + { + "epoch": 0.204296875, + "grad_norm": 0.2236049324274063, + "learning_rate": 0.0004648670176962957, + "loss": 1.9239, + "step": 2092 + }, + { + "epoch": 0.20439453125, + "grad_norm": 0.20630759000778198, + "learning_rate": 0.00046482885110589727, + "loss": 1.9762, + "step": 2093 + }, + { + "epoch": 0.2044921875, + "grad_norm": 0.25111427903175354, + "learning_rate": 0.0004647906655532784, + "loss": 1.9006, + "step": 2094 + }, + { + "epoch": 0.20458984375, + "grad_norm": 0.2501910626888275, + "learning_rate": 0.0004647524610422535, + "loss": 1.9586, + "step": 2095 + }, + { + "epoch": 0.2046875, + "grad_norm": 0.27904069423675537, + "learning_rate": 0.00046471423757663867, + "loss": 1.8846, + "step": 2096 + }, + { + "epoch": 0.20478515625, + "grad_norm": 0.24683643877506256, + "learning_rate": 0.0004646759951602523, + "loss": 1.8941, + "step": 2097 + }, + { + "epoch": 0.2048828125, + "grad_norm": 0.21275731921195984, + "learning_rate": 0.00046463773379691424, + "loss": 1.9168, + "step": 2098 + }, + { + "epoch": 0.20498046875, + "grad_norm": 0.2217152714729309, + "learning_rate": 0.0004645994534904468, + "loss": 1.8554, + "step": 2099 + }, + { + "epoch": 0.205078125, + "grad_norm": 0.20951475203037262, + "learning_rate": 0.0004645611542446736, + "loss": 1.9226, + "step": 2100 + }, + { + "epoch": 0.20517578125, + "grad_norm": 0.1962258368730545, + "learning_rate": 0.00046452283606342046, + "loss": 1.9244, + "step": 2101 + }, + { + "epoch": 0.2052734375, + "grad_norm": 0.25039345026016235, + "learning_rate": 0.000464484498950515, + "loss": 1.9237, + "step": 2102 + }, + { + "epoch": 0.20537109375, + "grad_norm": 0.2845058739185333, + "learning_rate": 0.00046444614290978674, + "loss": 1.8962, + "step": 2103 + }, + { + "epoch": 0.20546875, + "grad_norm": 0.22429201006889343, + "learning_rate": 0.0004644077679450672, + "loss": 1.9632, + "step": 2104 + }, + { + "epoch": 0.20556640625, + "grad_norm": 0.20088529586791992, + "learning_rate": 0.00046436937406018963, + "loss": 1.9211, + "step": 2105 + }, + { + "epoch": 0.2056640625, + "grad_norm": 0.2091323882341385, + "learning_rate": 0.0004643309612589893, + "loss": 1.913, + "step": 2106 + }, + { + "epoch": 0.20576171875, + "grad_norm": 0.24969804286956787, + "learning_rate": 0.00046429252954530316, + "loss": 1.8726, + "step": 2107 + }, + { + "epoch": 0.205859375, + "grad_norm": 0.30802738666534424, + "learning_rate": 0.00046425407892297025, + "loss": 1.9256, + "step": 2108 + }, + { + "epoch": 0.20595703125, + "grad_norm": 0.27234166860580444, + "learning_rate": 0.0004642156093958317, + "loss": 1.9548, + "step": 2109 + }, + { + "epoch": 0.2060546875, + "grad_norm": 0.19717903435230255, + "learning_rate": 0.00046417712096772994, + "loss": 1.9694, + "step": 2110 + }, + { + "epoch": 0.20615234375, + "grad_norm": 0.3146752715110779, + "learning_rate": 0.0004641386136425098, + "loss": 1.922, + "step": 2111 + }, + { + "epoch": 0.20625, + "grad_norm": 0.29423826932907104, + "learning_rate": 0.0004641000874240178, + "loss": 1.8961, + "step": 2112 + }, + { + "epoch": 0.20634765625, + "grad_norm": 0.23182527720928192, + "learning_rate": 0.0004640615423161022, + "loss": 1.9062, + "step": 2113 + }, + { + "epoch": 0.2064453125, + "grad_norm": 0.3178667426109314, + "learning_rate": 0.00046402297832261354, + "loss": 1.9098, + "step": 2114 + }, + { + "epoch": 0.20654296875, + "grad_norm": 0.3038199841976166, + "learning_rate": 0.000463984395447404, + "loss": 1.8738, + "step": 2115 + }, + { + "epoch": 0.206640625, + "grad_norm": 0.29474422335624695, + "learning_rate": 0.00046394579369432755, + "loss": 1.8506, + "step": 2116 + }, + { + "epoch": 0.20673828125, + "grad_norm": 0.30578935146331787, + "learning_rate": 0.00046390717306724023, + "loss": 1.8775, + "step": 2117 + }, + { + "epoch": 0.2068359375, + "grad_norm": 0.27422910928726196, + "learning_rate": 0.0004638685335699999, + "loss": 1.9101, + "step": 2118 + }, + { + "epoch": 0.20693359375, + "grad_norm": 0.2473389059305191, + "learning_rate": 0.00046382987520646624, + "loss": 1.9342, + "step": 2119 + }, + { + "epoch": 0.20703125, + "grad_norm": 0.2475956827402115, + "learning_rate": 0.00046379119798050094, + "loss": 1.9242, + "step": 2120 + }, + { + "epoch": 0.20712890625, + "grad_norm": 0.2836126983165741, + "learning_rate": 0.0004637525018959675, + "loss": 1.9263, + "step": 2121 + }, + { + "epoch": 0.2072265625, + "grad_norm": 0.20722109079360962, + "learning_rate": 0.00046371378695673137, + "loss": 1.9077, + "step": 2122 + }, + { + "epoch": 0.20732421875, + "grad_norm": 0.2630356550216675, + "learning_rate": 0.00046367505316665975, + "loss": 1.9109, + "step": 2123 + }, + { + "epoch": 0.207421875, + "grad_norm": 0.26636844873428345, + "learning_rate": 0.0004636363005296218, + "loss": 1.9011, + "step": 2124 + }, + { + "epoch": 0.20751953125, + "grad_norm": 0.3061290681362152, + "learning_rate": 0.0004635975290494887, + "loss": 1.8875, + "step": 2125 + }, + { + "epoch": 0.2076171875, + "grad_norm": 0.2455659955739975, + "learning_rate": 0.00046355873873013315, + "loss": 1.8745, + "step": 2126 + }, + { + "epoch": 0.20771484375, + "grad_norm": 0.226226806640625, + "learning_rate": 0.0004635199295754301, + "loss": 1.9186, + "step": 2127 + }, + { + "epoch": 0.2078125, + "grad_norm": 0.27755168080329895, + "learning_rate": 0.0004634811015892563, + "loss": 1.878, + "step": 2128 + }, + { + "epoch": 0.20791015625, + "grad_norm": 0.24024835228919983, + "learning_rate": 0.00046344225477549015, + "loss": 1.8886, + "step": 2129 + }, + { + "epoch": 0.2080078125, + "grad_norm": 0.2625919282436371, + "learning_rate": 0.0004634033891380122, + "loss": 1.8997, + "step": 2130 + }, + { + "epoch": 0.20810546875, + "grad_norm": 0.2913138270378113, + "learning_rate": 0.00046336450468070485, + "loss": 1.8711, + "step": 2131 + }, + { + "epoch": 0.208203125, + "grad_norm": 0.24820668995380402, + "learning_rate": 0.00046332560140745216, + "loss": 1.9243, + "step": 2132 + }, + { + "epoch": 0.20830078125, + "grad_norm": 0.2551608085632324, + "learning_rate": 0.00046328667932214035, + "loss": 1.8974, + "step": 2133 + }, + { + "epoch": 0.2083984375, + "grad_norm": 0.28112420439720154, + "learning_rate": 0.0004632477384286574, + "loss": 1.9122, + "step": 2134 + }, + { + "epoch": 0.20849609375, + "grad_norm": 0.23654119670391083, + "learning_rate": 0.0004632087787308931, + "loss": 1.93, + "step": 2135 + }, + { + "epoch": 0.20859375, + "grad_norm": 0.26119500398635864, + "learning_rate": 0.00046316980023273905, + "loss": 1.8936, + "step": 2136 + }, + { + "epoch": 0.20869140625, + "grad_norm": 0.26187509298324585, + "learning_rate": 0.0004631308029380891, + "loss": 1.9257, + "step": 2137 + }, + { + "epoch": 0.2087890625, + "grad_norm": 0.26531320810317993, + "learning_rate": 0.00046309178685083856, + "loss": 1.9203, + "step": 2138 + }, + { + "epoch": 0.20888671875, + "grad_norm": 0.28738391399383545, + "learning_rate": 0.00046305275197488496, + "loss": 1.9092, + "step": 2139 + }, + { + "epoch": 0.208984375, + "grad_norm": 0.2203925997018814, + "learning_rate": 0.0004630136983141273, + "loss": 1.9086, + "step": 2140 + }, + { + "epoch": 0.20908203125, + "grad_norm": 0.337566077709198, + "learning_rate": 0.0004629746258724669, + "loss": 1.896, + "step": 2141 + }, + { + "epoch": 0.2091796875, + "grad_norm": 0.22496585547924042, + "learning_rate": 0.0004629355346538067, + "loss": 1.9216, + "step": 2142 + }, + { + "epoch": 0.20927734375, + "grad_norm": 0.22534677386283875, + "learning_rate": 0.0004628964246620516, + "loss": 1.9074, + "step": 2143 + }, + { + "epoch": 0.209375, + "grad_norm": 0.24914197623729706, + "learning_rate": 0.0004628572959011082, + "loss": 1.9033, + "step": 2144 + }, + { + "epoch": 0.20947265625, + "grad_norm": 0.20345696806907654, + "learning_rate": 0.0004628181483748852, + "loss": 1.9218, + "step": 2145 + }, + { + "epoch": 0.2095703125, + "grad_norm": 0.22977322340011597, + "learning_rate": 0.00046277898208729306, + "loss": 1.8171, + "step": 2146 + }, + { + "epoch": 0.20966796875, + "grad_norm": 0.24085818231105804, + "learning_rate": 0.0004627397970422443, + "loss": 1.867, + "step": 2147 + }, + { + "epoch": 0.209765625, + "grad_norm": 0.26558682322502136, + "learning_rate": 0.00046270059324365296, + "loss": 1.9171, + "step": 2148 + }, + { + "epoch": 0.20986328125, + "grad_norm": 0.27644988894462585, + "learning_rate": 0.00046266137069543524, + "loss": 1.908, + "step": 2149 + }, + { + "epoch": 0.2099609375, + "grad_norm": 0.2609714865684509, + "learning_rate": 0.00046262212940150914, + "loss": 1.9649, + "step": 2150 + }, + { + "epoch": 0.21005859375, + "grad_norm": 0.27116766571998596, + "learning_rate": 0.00046258286936579453, + "loss": 1.9123, + "step": 2151 + }, + { + "epoch": 0.21015625, + "grad_norm": 0.21386384963989258, + "learning_rate": 0.0004625435905922131, + "loss": 1.9453, + "step": 2152 + }, + { + "epoch": 0.21025390625, + "grad_norm": 0.21030083298683167, + "learning_rate": 0.00046250429308468833, + "loss": 1.9349, + "step": 2153 + }, + { + "epoch": 0.2103515625, + "grad_norm": 0.267782062292099, + "learning_rate": 0.0004624649768471459, + "loss": 1.9266, + "step": 2154 + }, + { + "epoch": 0.21044921875, + "grad_norm": 0.2850502133369446, + "learning_rate": 0.0004624256418835131, + "loss": 1.884, + "step": 2155 + }, + { + "epoch": 0.210546875, + "grad_norm": 0.3452877700328827, + "learning_rate": 0.0004623862881977191, + "loss": 1.9651, + "step": 2156 + }, + { + "epoch": 0.21064453125, + "grad_norm": 0.2617965638637543, + "learning_rate": 0.00046234691579369504, + "loss": 1.8952, + "step": 2157 + }, + { + "epoch": 0.2107421875, + "grad_norm": 0.2741365134716034, + "learning_rate": 0.00046230752467537375, + "loss": 1.8951, + "step": 2158 + }, + { + "epoch": 0.21083984375, + "grad_norm": 0.37253716588020325, + "learning_rate": 0.00046226811484669015, + "loss": 1.9242, + "step": 2159 + }, + { + "epoch": 0.2109375, + "grad_norm": 0.28392404317855835, + "learning_rate": 0.00046222868631158105, + "loss": 1.9025, + "step": 2160 + }, + { + "epoch": 0.21103515625, + "grad_norm": 0.2524137794971466, + "learning_rate": 0.00046218923907398473, + "loss": 1.9166, + "step": 2161 + }, + { + "epoch": 0.2111328125, + "grad_norm": 0.2966980040073395, + "learning_rate": 0.0004621497731378419, + "loss": 1.9083, + "step": 2162 + }, + { + "epoch": 0.21123046875, + "grad_norm": 0.28292515873908997, + "learning_rate": 0.0004621102885070946, + "loss": 1.9049, + "step": 2163 + }, + { + "epoch": 0.211328125, + "grad_norm": 0.2480284720659256, + "learning_rate": 0.00046207078518568717, + "loss": 1.9063, + "step": 2164 + }, + { + "epoch": 0.21142578125, + "grad_norm": 0.2493746429681778, + "learning_rate": 0.0004620312631775657, + "loss": 1.8636, + "step": 2165 + }, + { + "epoch": 0.2115234375, + "grad_norm": 0.19225762784481049, + "learning_rate": 0.0004619917224866779, + "loss": 1.8916, + "step": 2166 + }, + { + "epoch": 0.21162109375, + "grad_norm": 0.24277755618095398, + "learning_rate": 0.00046195216311697355, + "loss": 1.9265, + "step": 2167 + }, + { + "epoch": 0.21171875, + "grad_norm": 0.2281288206577301, + "learning_rate": 0.0004619125850724045, + "loss": 1.9044, + "step": 2168 + }, + { + "epoch": 0.21181640625, + "grad_norm": 0.24480199813842773, + "learning_rate": 0.000461872988356924, + "loss": 1.8734, + "step": 2169 + }, + { + "epoch": 0.2119140625, + "grad_norm": 0.2577948570251465, + "learning_rate": 0.0004618333729744876, + "loss": 1.9017, + "step": 2170 + }, + { + "epoch": 0.21201171875, + "grad_norm": 0.22131475806236267, + "learning_rate": 0.00046179373892905233, + "loss": 1.9071, + "step": 2171 + }, + { + "epoch": 0.212109375, + "grad_norm": 0.24380597472190857, + "learning_rate": 0.00046175408622457743, + "loss": 1.9291, + "step": 2172 + }, + { + "epoch": 0.21220703125, + "grad_norm": 0.25301221013069153, + "learning_rate": 0.0004617144148650238, + "loss": 1.8888, + "step": 2173 + }, + { + "epoch": 0.2123046875, + "grad_norm": 0.2644136846065521, + "learning_rate": 0.00046167472485435424, + "loss": 1.8995, + "step": 2174 + }, + { + "epoch": 0.21240234375, + "grad_norm": 0.2486433982849121, + "learning_rate": 0.0004616350161965335, + "loss": 1.9229, + "step": 2175 + }, + { + "epoch": 0.2125, + "grad_norm": 0.3275326192378998, + "learning_rate": 0.0004615952888955281, + "loss": 1.9174, + "step": 2176 + }, + { + "epoch": 0.21259765625, + "grad_norm": 0.22458401322364807, + "learning_rate": 0.00046155554295530636, + "loss": 1.8847, + "step": 2177 + }, + { + "epoch": 0.2126953125, + "grad_norm": 0.2412051111459732, + "learning_rate": 0.0004615157783798387, + "loss": 1.9229, + "step": 2178 + }, + { + "epoch": 0.21279296875, + "grad_norm": 0.21972574293613434, + "learning_rate": 0.00046147599517309707, + "loss": 1.9243, + "step": 2179 + }, + { + "epoch": 0.212890625, + "grad_norm": 0.2897114157676697, + "learning_rate": 0.0004614361933390555, + "loss": 1.9192, + "step": 2180 + }, + { + "epoch": 0.21298828125, + "grad_norm": 0.2781403660774231, + "learning_rate": 0.00046139637288169003, + "loss": 1.8995, + "step": 2181 + }, + { + "epoch": 0.2130859375, + "grad_norm": 0.2788412272930145, + "learning_rate": 0.00046135653380497814, + "loss": 1.9041, + "step": 2182 + }, + { + "epoch": 0.21318359375, + "grad_norm": 0.28828147053718567, + "learning_rate": 0.0004613166761128996, + "loss": 1.9016, + "step": 2183 + }, + { + "epoch": 0.21328125, + "grad_norm": 0.26774659752845764, + "learning_rate": 0.00046127679980943563, + "loss": 1.9703, + "step": 2184 + }, + { + "epoch": 0.21337890625, + "grad_norm": 0.2517338991165161, + "learning_rate": 0.00046123690489856966, + "loss": 1.8761, + "step": 2185 + }, + { + "epoch": 0.2134765625, + "grad_norm": 0.25602057576179504, + "learning_rate": 0.00046119699138428676, + "loss": 1.9538, + "step": 2186 + }, + { + "epoch": 0.21357421875, + "grad_norm": 0.20147521793842316, + "learning_rate": 0.0004611570592705741, + "loss": 1.8697, + "step": 2187 + }, + { + "epoch": 0.213671875, + "grad_norm": 0.22156640887260437, + "learning_rate": 0.0004611171085614203, + "loss": 1.8982, + "step": 2188 + }, + { + "epoch": 0.21376953125, + "grad_norm": 0.23269224166870117, + "learning_rate": 0.00046107713926081626, + "loss": 1.8925, + "step": 2189 + }, + { + "epoch": 0.2138671875, + "grad_norm": 0.21519561111927032, + "learning_rate": 0.00046103715137275447, + "loss": 1.8476, + "step": 2190 + }, + { + "epoch": 0.21396484375, + "grad_norm": 0.2535529136657715, + "learning_rate": 0.0004609971449012294, + "loss": 1.901, + "step": 2191 + }, + { + "epoch": 0.2140625, + "grad_norm": 0.24704895913600922, + "learning_rate": 0.0004609571198502374, + "loss": 1.8946, + "step": 2192 + }, + { + "epoch": 0.21416015625, + "grad_norm": 0.24865074455738068, + "learning_rate": 0.00046091707622377646, + "loss": 1.8576, + "step": 2193 + }, + { + "epoch": 0.2142578125, + "grad_norm": 0.2663979232311249, + "learning_rate": 0.00046087701402584683, + "loss": 1.89, + "step": 2194 + }, + { + "epoch": 0.21435546875, + "grad_norm": 0.2134150117635727, + "learning_rate": 0.00046083693326045004, + "loss": 1.9338, + "step": 2195 + }, + { + "epoch": 0.214453125, + "grad_norm": 0.20297788083553314, + "learning_rate": 0.00046079683393159004, + "loss": 1.9193, + "step": 2196 + }, + { + "epoch": 0.21455078125, + "grad_norm": 0.22827771306037903, + "learning_rate": 0.0004607567160432724, + "loss": 1.9089, + "step": 2197 + }, + { + "epoch": 0.2146484375, + "grad_norm": 0.2234564572572708, + "learning_rate": 0.0004607165795995044, + "loss": 1.9069, + "step": 2198 + }, + { + "epoch": 0.21474609375, + "grad_norm": 0.23381231725215912, + "learning_rate": 0.0004606764246042955, + "loss": 1.9059, + "step": 2199 + }, + { + "epoch": 0.21484375, + "grad_norm": 0.24767504632472992, + "learning_rate": 0.0004606362510616567, + "loss": 1.9448, + "step": 2200 + }, + { + "epoch": 0.21494140625, + "grad_norm": 0.23719260096549988, + "learning_rate": 0.00046059605897560095, + "loss": 1.8839, + "step": 2201 + }, + { + "epoch": 0.2150390625, + "grad_norm": 0.35658907890319824, + "learning_rate": 0.00046055584835014325, + "loss": 1.912, + "step": 2202 + }, + { + "epoch": 0.21513671875, + "grad_norm": 0.35971537232398987, + "learning_rate": 0.0004605156191893001, + "loss": 1.9476, + "step": 2203 + }, + { + "epoch": 0.215234375, + "grad_norm": 0.242756187915802, + "learning_rate": 0.0004604753714970901, + "loss": 1.8945, + "step": 2204 + }, + { + "epoch": 0.21533203125, + "grad_norm": 0.299610435962677, + "learning_rate": 0.0004604351052775337, + "loss": 1.9209, + "step": 2205 + }, + { + "epoch": 0.2154296875, + "grad_norm": 0.21338649094104767, + "learning_rate": 0.000460394820534653, + "loss": 1.8975, + "step": 2206 + }, + { + "epoch": 0.21552734375, + "grad_norm": 0.2629339098930359, + "learning_rate": 0.00046035451727247226, + "loss": 1.8938, + "step": 2207 + }, + { + "epoch": 0.215625, + "grad_norm": 0.2574734091758728, + "learning_rate": 0.00046031419549501733, + "loss": 1.9167, + "step": 2208 + }, + { + "epoch": 0.21572265625, + "grad_norm": 0.20906251668930054, + "learning_rate": 0.000460273855206316, + "loss": 1.8635, + "step": 2209 + }, + { + "epoch": 0.2158203125, + "grad_norm": 0.265655517578125, + "learning_rate": 0.00046023349641039786, + "loss": 1.9223, + "step": 2210 + }, + { + "epoch": 0.21591796875, + "grad_norm": 0.308352530002594, + "learning_rate": 0.0004601931191112945, + "loss": 1.9127, + "step": 2211 + }, + { + "epoch": 0.216015625, + "grad_norm": 0.26350700855255127, + "learning_rate": 0.0004601527233130392, + "loss": 1.8908, + "step": 2212 + }, + { + "epoch": 0.21611328125, + "grad_norm": 0.28734543919563293, + "learning_rate": 0.00046011230901966716, + "loss": 1.9426, + "step": 2213 + }, + { + "epoch": 0.2162109375, + "grad_norm": 0.24283497035503387, + "learning_rate": 0.0004600718762352154, + "loss": 1.915, + "step": 2214 + }, + { + "epoch": 0.21630859375, + "grad_norm": 0.24215170741081238, + "learning_rate": 0.00046003142496372275, + "loss": 1.9154, + "step": 2215 + }, + { + "epoch": 0.21640625, + "grad_norm": 0.25598227977752686, + "learning_rate": 0.0004599909552092301, + "loss": 1.9225, + "step": 2216 + }, + { + "epoch": 0.21650390625, + "grad_norm": 0.2031700313091278, + "learning_rate": 0.00045995046697577975, + "loss": 1.8965, + "step": 2217 + }, + { + "epoch": 0.2166015625, + "grad_norm": 0.28480711579322815, + "learning_rate": 0.0004599099602674163, + "loss": 1.9212, + "step": 2218 + }, + { + "epoch": 0.21669921875, + "grad_norm": 0.22612528502941132, + "learning_rate": 0.000459869435088186, + "loss": 1.9439, + "step": 2219 + }, + { + "epoch": 0.216796875, + "grad_norm": 0.22312787175178528, + "learning_rate": 0.0004598288914421369, + "loss": 1.9238, + "step": 2220 + }, + { + "epoch": 0.21689453125, + "grad_norm": 0.24936683475971222, + "learning_rate": 0.000459788329333319, + "loss": 1.8923, + "step": 2221 + }, + { + "epoch": 0.2169921875, + "grad_norm": 0.2038477212190628, + "learning_rate": 0.00045974774876578406, + "loss": 1.8993, + "step": 2222 + }, + { + "epoch": 0.21708984375, + "grad_norm": 0.20470038056373596, + "learning_rate": 0.00045970714974358576, + "loss": 1.901, + "step": 2223 + }, + { + "epoch": 0.2171875, + "grad_norm": 0.20012640953063965, + "learning_rate": 0.00045966653227077955, + "loss": 1.8809, + "step": 2224 + }, + { + "epoch": 0.21728515625, + "grad_norm": 0.1954420655965805, + "learning_rate": 0.0004596258963514228, + "loss": 1.9468, + "step": 2225 + }, + { + "epoch": 0.2173828125, + "grad_norm": 0.19190649688243866, + "learning_rate": 0.00045958524198957463, + "loss": 1.8548, + "step": 2226 + }, + { + "epoch": 0.21748046875, + "grad_norm": 0.2360774725675583, + "learning_rate": 0.000459544569189296, + "loss": 1.8779, + "step": 2227 + }, + { + "epoch": 0.217578125, + "grad_norm": 0.24308523535728455, + "learning_rate": 0.0004595038779546499, + "loss": 1.9122, + "step": 2228 + }, + { + "epoch": 0.21767578125, + "grad_norm": 0.31237441301345825, + "learning_rate": 0.00045946316828970093, + "loss": 1.9101, + "step": 2229 + }, + { + "epoch": 0.2177734375, + "grad_norm": 0.26211169362068176, + "learning_rate": 0.00045942244019851557, + "loss": 1.8804, + "step": 2230 + }, + { + "epoch": 0.21787109375, + "grad_norm": 0.2568318843841553, + "learning_rate": 0.0004593816936851623, + "loss": 1.8625, + "step": 2231 + }, + { + "epoch": 0.21796875, + "grad_norm": 0.2667257487773895, + "learning_rate": 0.00045934092875371134, + "loss": 1.8856, + "step": 2232 + }, + { + "epoch": 0.21806640625, + "grad_norm": 0.24612100422382355, + "learning_rate": 0.0004593001454082347, + "loss": 1.8921, + "step": 2233 + }, + { + "epoch": 0.2181640625, + "grad_norm": 0.2511996626853943, + "learning_rate": 0.0004592593436528063, + "loss": 1.8955, + "step": 2234 + }, + { + "epoch": 0.21826171875, + "grad_norm": 0.2945897877216339, + "learning_rate": 0.0004592185234915019, + "loss": 1.8911, + "step": 2235 + }, + { + "epoch": 0.218359375, + "grad_norm": 0.34499260783195496, + "learning_rate": 0.00045917768492839895, + "loss": 1.9234, + "step": 2236 + }, + { + "epoch": 0.21845703125, + "grad_norm": 0.20355650782585144, + "learning_rate": 0.000459136827967577, + "loss": 1.8892, + "step": 2237 + }, + { + "epoch": 0.2185546875, + "grad_norm": 0.31010961532592773, + "learning_rate": 0.0004590959526131172, + "loss": 1.8915, + "step": 2238 + }, + { + "epoch": 0.21865234375, + "grad_norm": 0.3139292001724243, + "learning_rate": 0.0004590550588691028, + "loss": 1.9084, + "step": 2239 + }, + { + "epoch": 0.21875, + "grad_norm": 0.2442852258682251, + "learning_rate": 0.00045901414673961844, + "loss": 1.8947, + "step": 2240 + }, + { + "epoch": 0.21884765625, + "grad_norm": 0.27955976128578186, + "learning_rate": 0.00045897321622875105, + "loss": 1.8675, + "step": 2241 + }, + { + "epoch": 0.2189453125, + "grad_norm": 0.1819334477186203, + "learning_rate": 0.00045893226734058923, + "loss": 1.8999, + "step": 2242 + }, + { + "epoch": 0.21904296875, + "grad_norm": 0.24583211541175842, + "learning_rate": 0.00045889130007922347, + "loss": 1.8846, + "step": 2243 + }, + { + "epoch": 0.219140625, + "grad_norm": 0.2537577450275421, + "learning_rate": 0.00045885031444874597, + "loss": 1.9059, + "step": 2244 + }, + { + "epoch": 0.21923828125, + "grad_norm": 0.22242943942546844, + "learning_rate": 0.00045880931045325074, + "loss": 1.8862, + "step": 2245 + }, + { + "epoch": 0.2193359375, + "grad_norm": 0.2518673241138458, + "learning_rate": 0.0004587682880968338, + "loss": 1.9075, + "step": 2246 + }, + { + "epoch": 0.21943359375, + "grad_norm": 0.2151048630475998, + "learning_rate": 0.00045872724738359297, + "loss": 1.9299, + "step": 2247 + }, + { + "epoch": 0.21953125, + "grad_norm": 0.24207067489624023, + "learning_rate": 0.00045868618831762775, + "loss": 1.8986, + "step": 2248 + }, + { + "epoch": 0.21962890625, + "grad_norm": 0.2483375072479248, + "learning_rate": 0.0004586451109030397, + "loss": 1.9244, + "step": 2249 + }, + { + "epoch": 0.2197265625, + "grad_norm": 0.19953475892543793, + "learning_rate": 0.000458604015143932, + "loss": 1.9284, + "step": 2250 + }, + { + "epoch": 0.21982421875, + "grad_norm": 0.20768243074417114, + "learning_rate": 0.00045856290104440974, + "loss": 1.8846, + "step": 2251 + }, + { + "epoch": 0.219921875, + "grad_norm": 0.21332871913909912, + "learning_rate": 0.0004585217686085799, + "loss": 1.8777, + "step": 2252 + }, + { + "epoch": 0.22001953125, + "grad_norm": 0.22202058136463165, + "learning_rate": 0.0004584806178405512, + "loss": 1.8778, + "step": 2253 + }, + { + "epoch": 0.2201171875, + "grad_norm": 0.22092659771442413, + "learning_rate": 0.00045843944874443435, + "loss": 1.9006, + "step": 2254 + }, + { + "epoch": 0.22021484375, + "grad_norm": 0.2076462358236313, + "learning_rate": 0.0004583982613243416, + "loss": 1.8781, + "step": 2255 + }, + { + "epoch": 0.2203125, + "grad_norm": 0.26965686678886414, + "learning_rate": 0.0004583570555843874, + "loss": 1.8818, + "step": 2256 + }, + { + "epoch": 0.22041015625, + "grad_norm": 0.23709794878959656, + "learning_rate": 0.0004583158315286877, + "loss": 1.799, + "step": 2257 + }, + { + "epoch": 0.2205078125, + "grad_norm": 0.21832555532455444, + "learning_rate": 0.00045827458916136043, + "loss": 1.9321, + "step": 2258 + }, + { + "epoch": 0.22060546875, + "grad_norm": 0.268564909696579, + "learning_rate": 0.0004582333284865254, + "loss": 1.9043, + "step": 2259 + }, + { + "epoch": 0.220703125, + "grad_norm": 0.2340027242898941, + "learning_rate": 0.00045819204950830426, + "loss": 1.8435, + "step": 2260 + }, + { + "epoch": 0.22080078125, + "grad_norm": 0.2653496563434601, + "learning_rate": 0.00045815075223082016, + "loss": 1.8857, + "step": 2261 + }, + { + "epoch": 0.2208984375, + "grad_norm": 0.2956698536872864, + "learning_rate": 0.00045810943665819853, + "loss": 1.9683, + "step": 2262 + }, + { + "epoch": 0.22099609375, + "grad_norm": 0.2561390697956085, + "learning_rate": 0.0004580681027945663, + "loss": 1.9258, + "step": 2263 + }, + { + "epoch": 0.22109375, + "grad_norm": 0.2162081003189087, + "learning_rate": 0.00045802675064405266, + "loss": 1.9093, + "step": 2264 + }, + { + "epoch": 0.22119140625, + "grad_norm": 0.2608626186847687, + "learning_rate": 0.0004579853802107879, + "loss": 1.8742, + "step": 2265 + }, + { + "epoch": 0.2212890625, + "grad_norm": 0.2352752387523651, + "learning_rate": 0.0004579439914989049, + "loss": 1.9224, + "step": 2266 + }, + { + "epoch": 0.22138671875, + "grad_norm": 0.20996929705142975, + "learning_rate": 0.00045790258451253776, + "loss": 1.9081, + "step": 2267 + }, + { + "epoch": 0.221484375, + "grad_norm": 0.2689194083213806, + "learning_rate": 0.00045786115925582283, + "loss": 1.928, + "step": 2268 + }, + { + "epoch": 0.22158203125, + "grad_norm": 0.27555009722709656, + "learning_rate": 0.0004578197157328981, + "loss": 1.8619, + "step": 2269 + }, + { + "epoch": 0.2216796875, + "grad_norm": 0.23493286967277527, + "learning_rate": 0.00045777825394790344, + "loss": 1.9058, + "step": 2270 + }, + { + "epoch": 0.22177734375, + "grad_norm": 0.18501000106334686, + "learning_rate": 0.0004577367739049804, + "loss": 1.935, + "step": 2271 + }, + { + "epoch": 0.221875, + "grad_norm": 0.2200579047203064, + "learning_rate": 0.00045769527560827263, + "loss": 1.9219, + "step": 2272 + }, + { + "epoch": 0.22197265625, + "grad_norm": 0.24917781352996826, + "learning_rate": 0.0004576537590619254, + "loss": 1.8729, + "step": 2273 + }, + { + "epoch": 0.2220703125, + "grad_norm": 0.2548181116580963, + "learning_rate": 0.0004576122242700857, + "loss": 1.8637, + "step": 2274 + }, + { + "epoch": 0.22216796875, + "grad_norm": 0.27391573786735535, + "learning_rate": 0.0004575706712369027, + "loss": 1.9145, + "step": 2275 + }, + { + "epoch": 0.222265625, + "grad_norm": 0.21581903100013733, + "learning_rate": 0.00045752909996652695, + "loss": 1.9498, + "step": 2276 + }, + { + "epoch": 0.22236328125, + "grad_norm": 0.2738785147666931, + "learning_rate": 0.00045748751046311125, + "loss": 1.9297, + "step": 2277 + }, + { + "epoch": 0.2224609375, + "grad_norm": 0.2592880427837372, + "learning_rate": 0.00045744590273080987, + "loss": 1.9323, + "step": 2278 + }, + { + "epoch": 0.22255859375, + "grad_norm": 0.21176159381866455, + "learning_rate": 0.00045740427677377926, + "loss": 1.9299, + "step": 2279 + }, + { + "epoch": 0.22265625, + "grad_norm": 0.2628064751625061, + "learning_rate": 0.00045736263259617727, + "loss": 1.9047, + "step": 2280 + }, + { + "epoch": 0.22275390625, + "grad_norm": 0.199008047580719, + "learning_rate": 0.00045732097020216393, + "loss": 1.8896, + "step": 2281 + }, + { + "epoch": 0.2228515625, + "grad_norm": 0.2588101923465729, + "learning_rate": 0.00045727928959590086, + "loss": 1.9043, + "step": 2282 + }, + { + "epoch": 0.22294921875, + "grad_norm": 0.2765114903450012, + "learning_rate": 0.00045723759078155165, + "loss": 1.8701, + "step": 2283 + }, + { + "epoch": 0.223046875, + "grad_norm": 0.22020654380321503, + "learning_rate": 0.0004571958737632815, + "loss": 1.842, + "step": 2284 + }, + { + "epoch": 0.22314453125, + "grad_norm": 0.25397929549217224, + "learning_rate": 0.0004571541385452577, + "loss": 1.9356, + "step": 2285 + }, + { + "epoch": 0.2232421875, + "grad_norm": 0.3181978464126587, + "learning_rate": 0.0004571123851316492, + "loss": 1.8812, + "step": 2286 + }, + { + "epoch": 0.22333984375, + "grad_norm": 0.24803847074508667, + "learning_rate": 0.0004570706135266268, + "loss": 1.9017, + "step": 2287 + }, + { + "epoch": 0.2234375, + "grad_norm": 0.2591163218021393, + "learning_rate": 0.00045702882373436317, + "loss": 1.8597, + "step": 2288 + }, + { + "epoch": 0.22353515625, + "grad_norm": 0.2701427936553955, + "learning_rate": 0.00045698701575903265, + "loss": 1.9202, + "step": 2289 + }, + { + "epoch": 0.2236328125, + "grad_norm": 0.259224534034729, + "learning_rate": 0.00045694518960481145, + "loss": 1.8726, + "step": 2290 + }, + { + "epoch": 0.22373046875, + "grad_norm": 0.2605198919773102, + "learning_rate": 0.00045690334527587786, + "loss": 1.8942, + "step": 2291 + }, + { + "epoch": 0.223828125, + "grad_norm": 0.2564171254634857, + "learning_rate": 0.00045686148277641143, + "loss": 1.8893, + "step": 2292 + }, + { + "epoch": 0.22392578125, + "grad_norm": 0.2904933989048004, + "learning_rate": 0.0004568196021105941, + "loss": 1.8712, + "step": 2293 + }, + { + "epoch": 0.2240234375, + "grad_norm": 0.19899466633796692, + "learning_rate": 0.0004567777032826092, + "loss": 1.882, + "step": 2294 + }, + { + "epoch": 0.22412109375, + "grad_norm": 0.3275931477546692, + "learning_rate": 0.00045673578629664227, + "loss": 1.9256, + "step": 2295 + }, + { + "epoch": 0.22421875, + "grad_norm": 0.25155848264694214, + "learning_rate": 0.0004566938511568802, + "loss": 1.8914, + "step": 2296 + }, + { + "epoch": 0.22431640625, + "grad_norm": 0.2259829044342041, + "learning_rate": 0.00045665189786751214, + "loss": 1.9638, + "step": 2297 + }, + { + "epoch": 0.2244140625, + "grad_norm": 0.21916180849075317, + "learning_rate": 0.0004566099264327288, + "loss": 1.8701, + "step": 2298 + }, + { + "epoch": 0.22451171875, + "grad_norm": 0.2307112216949463, + "learning_rate": 0.0004565679368567227, + "loss": 1.9039, + "step": 2299 + }, + { + "epoch": 0.224609375, + "grad_norm": 0.25406181812286377, + "learning_rate": 0.00045652592914368826, + "loss": 1.8874, + "step": 2300 + }, + { + "epoch": 0.22470703125, + "grad_norm": 0.23334388434886932, + "learning_rate": 0.0004564839032978216, + "loss": 1.9329, + "step": 2301 + }, + { + "epoch": 0.2248046875, + "grad_norm": 0.312667578458786, + "learning_rate": 0.0004564418593233209, + "loss": 1.8981, + "step": 2302 + }, + { + "epoch": 0.22490234375, + "grad_norm": 0.21629926562309265, + "learning_rate": 0.00045639979722438586, + "loss": 1.9121, + "step": 2303 + }, + { + "epoch": 0.225, + "grad_norm": 0.2639700174331665, + "learning_rate": 0.0004563577170052182, + "loss": 1.8917, + "step": 2304 + }, + { + "epoch": 0.22509765625, + "grad_norm": 0.28482767939567566, + "learning_rate": 0.00045631561867002114, + "loss": 1.8615, + "step": 2305 + }, + { + "epoch": 0.2251953125, + "grad_norm": 0.21128278970718384, + "learning_rate": 0.0004562735022230002, + "loss": 1.8742, + "step": 2306 + }, + { + "epoch": 0.22529296875, + "grad_norm": 0.22739477455615997, + "learning_rate": 0.00045623136766836226, + "loss": 1.8953, + "step": 2307 + }, + { + "epoch": 0.225390625, + "grad_norm": 0.2244565635919571, + "learning_rate": 0.0004561892150103163, + "loss": 1.8773, + "step": 2308 + }, + { + "epoch": 0.22548828125, + "grad_norm": 0.24845881760120392, + "learning_rate": 0.00045614704425307295, + "loss": 1.9182, + "step": 2309 + }, + { + "epoch": 0.2255859375, + "grad_norm": 0.24752497673034668, + "learning_rate": 0.0004561048554008447, + "loss": 1.8817, + "step": 2310 + }, + { + "epoch": 0.22568359375, + "grad_norm": 0.20452168583869934, + "learning_rate": 0.00045606264845784577, + "loss": 1.8662, + "step": 2311 + }, + { + "epoch": 0.22578125, + "grad_norm": 0.22137118875980377, + "learning_rate": 0.00045602042342829244, + "loss": 1.8966, + "step": 2312 + }, + { + "epoch": 0.22587890625, + "grad_norm": 0.2197723239660263, + "learning_rate": 0.00045597818031640244, + "loss": 1.9152, + "step": 2313 + }, + { + "epoch": 0.2259765625, + "grad_norm": 0.19033396244049072, + "learning_rate": 0.0004559359191263955, + "loss": 1.92, + "step": 2314 + }, + { + "epoch": 0.22607421875, + "grad_norm": 0.21937741339206696, + "learning_rate": 0.0004558936398624933, + "loss": 1.9554, + "step": 2315 + }, + { + "epoch": 0.226171875, + "grad_norm": 0.20096588134765625, + "learning_rate": 0.000455851342528919, + "loss": 1.9401, + "step": 2316 + }, + { + "epoch": 0.22626953125, + "grad_norm": 0.22804000973701477, + "learning_rate": 0.00045580902712989783, + "loss": 1.8753, + "step": 2317 + }, + { + "epoch": 0.2263671875, + "grad_norm": 0.2619505226612091, + "learning_rate": 0.0004557666936696566, + "loss": 1.8913, + "step": 2318 + }, + { + "epoch": 0.22646484375, + "grad_norm": 0.2752479016780853, + "learning_rate": 0.0004557243421524241, + "loss": 1.8744, + "step": 2319 + }, + { + "epoch": 0.2265625, + "grad_norm": 0.2984503209590912, + "learning_rate": 0.0004556819725824309, + "loss": 1.8951, + "step": 2320 + }, + { + "epoch": 0.22666015625, + "grad_norm": 0.22869320213794708, + "learning_rate": 0.0004556395849639094, + "loss": 1.9453, + "step": 2321 + }, + { + "epoch": 0.2267578125, + "grad_norm": 0.2314172238111496, + "learning_rate": 0.0004555971793010937, + "loss": 1.8549, + "step": 2322 + }, + { + "epoch": 0.22685546875, + "grad_norm": 0.2702145278453827, + "learning_rate": 0.0004555547555982197, + "loss": 1.8916, + "step": 2323 + }, + { + "epoch": 0.226953125, + "grad_norm": 0.29122287034988403, + "learning_rate": 0.00045551231385952516, + "loss": 1.9072, + "step": 2324 + }, + { + "epoch": 0.22705078125, + "grad_norm": 0.26239168643951416, + "learning_rate": 0.0004554698540892497, + "loss": 1.9243, + "step": 2325 + }, + { + "epoch": 0.2271484375, + "grad_norm": 0.25048989057540894, + "learning_rate": 0.0004554273762916346, + "loss": 1.8799, + "step": 2326 + }, + { + "epoch": 0.22724609375, + "grad_norm": 0.21093380451202393, + "learning_rate": 0.0004553848804709231, + "loss": 1.928, + "step": 2327 + }, + { + "epoch": 0.22734375, + "grad_norm": 0.24841107428073883, + "learning_rate": 0.00045534236663136, + "loss": 1.867, + "step": 2328 + }, + { + "epoch": 0.22744140625, + "grad_norm": 0.3298115134239197, + "learning_rate": 0.0004552998347771923, + "loss": 1.9387, + "step": 2329 + }, + { + "epoch": 0.2275390625, + "grad_norm": 0.2536830008029938, + "learning_rate": 0.0004552572849126683, + "loss": 1.8998, + "step": 2330 + }, + { + "epoch": 0.22763671875, + "grad_norm": 0.25619152188301086, + "learning_rate": 0.0004552147170420386, + "loss": 1.9069, + "step": 2331 + }, + { + "epoch": 0.227734375, + "grad_norm": 0.21483200788497925, + "learning_rate": 0.00045517213116955514, + "loss": 1.8838, + "step": 2332 + }, + { + "epoch": 0.22783203125, + "grad_norm": 0.24529071152210236, + "learning_rate": 0.00045512952729947196, + "loss": 1.8954, + "step": 2333 + }, + { + "epoch": 0.2279296875, + "grad_norm": 0.2747213840484619, + "learning_rate": 0.0004550869054360448, + "loss": 1.8707, + "step": 2334 + }, + { + "epoch": 0.22802734375, + "grad_norm": 0.24370868504047394, + "learning_rate": 0.00045504426558353123, + "loss": 1.8723, + "step": 2335 + }, + { + "epoch": 0.228125, + "grad_norm": 0.2064986526966095, + "learning_rate": 0.0004550016077461906, + "loss": 1.893, + "step": 2336 + }, + { + "epoch": 0.22822265625, + "grad_norm": 0.28048041462898254, + "learning_rate": 0.000454958931928284, + "loss": 1.8731, + "step": 2337 + }, + { + "epoch": 0.2283203125, + "grad_norm": 0.2515961527824402, + "learning_rate": 0.0004549162381340744, + "loss": 1.9322, + "step": 2338 + }, + { + "epoch": 0.22841796875, + "grad_norm": 0.21575289964675903, + "learning_rate": 0.0004548735263678266, + "loss": 1.8622, + "step": 2339 + }, + { + "epoch": 0.228515625, + "grad_norm": 0.22191932797431946, + "learning_rate": 0.00045483079663380686, + "loss": 1.9339, + "step": 2340 + }, + { + "epoch": 0.22861328125, + "grad_norm": 0.27168577909469604, + "learning_rate": 0.0004547880489362838, + "loss": 1.8946, + "step": 2341 + }, + { + "epoch": 0.2287109375, + "grad_norm": 0.2825971841812134, + "learning_rate": 0.0004547452832795275, + "loss": 1.9222, + "step": 2342 + }, + { + "epoch": 0.22880859375, + "grad_norm": 0.2461196780204773, + "learning_rate": 0.00045470249966780976, + "loss": 1.8803, + "step": 2343 + }, + { + "epoch": 0.22890625, + "grad_norm": 0.2580738067626953, + "learning_rate": 0.00045465969810540427, + "loss": 1.9044, + "step": 2344 + }, + { + "epoch": 0.22900390625, + "grad_norm": 0.22742857038974762, + "learning_rate": 0.0004546168785965866, + "loss": 1.8423, + "step": 2345 + }, + { + "epoch": 0.2291015625, + "grad_norm": 0.2607462406158447, + "learning_rate": 0.00045457404114563406, + "loss": 1.9238, + "step": 2346 + }, + { + "epoch": 0.22919921875, + "grad_norm": 0.21950286626815796, + "learning_rate": 0.00045453118575682565, + "loss": 1.8932, + "step": 2347 + }, + { + "epoch": 0.229296875, + "grad_norm": 0.2564532160758972, + "learning_rate": 0.0004544883124344423, + "loss": 1.8941, + "step": 2348 + }, + { + "epoch": 0.22939453125, + "grad_norm": 0.2837105393409729, + "learning_rate": 0.0004544454211827667, + "loss": 1.9385, + "step": 2349 + }, + { + "epoch": 0.2294921875, + "grad_norm": 0.268484503030777, + "learning_rate": 0.00045440251200608316, + "loss": 1.8984, + "step": 2350 + }, + { + "epoch": 0.22958984375, + "grad_norm": 0.3416820466518402, + "learning_rate": 0.0004543595849086782, + "loss": 1.7954, + "step": 2351 + }, + { + "epoch": 0.2296875, + "grad_norm": 0.3154846727848053, + "learning_rate": 0.00045431663989483955, + "loss": 1.8759, + "step": 2352 + }, + { + "epoch": 0.22978515625, + "grad_norm": 0.29342061281204224, + "learning_rate": 0.00045427367696885727, + "loss": 1.8975, + "step": 2353 + }, + { + "epoch": 0.2298828125, + "grad_norm": 0.27414578199386597, + "learning_rate": 0.00045423069613502275, + "loss": 1.8983, + "step": 2354 + }, + { + "epoch": 0.22998046875, + "grad_norm": 0.2255915254354477, + "learning_rate": 0.00045418769739762966, + "loss": 1.9371, + "step": 2355 + }, + { + "epoch": 0.230078125, + "grad_norm": 0.21754321455955505, + "learning_rate": 0.000454144680760973, + "loss": 1.8771, + "step": 2356 + }, + { + "epoch": 0.23017578125, + "grad_norm": 0.26277226209640503, + "learning_rate": 0.0004541016462293498, + "loss": 1.8672, + "step": 2357 + }, + { + "epoch": 0.2302734375, + "grad_norm": 0.22035324573516846, + "learning_rate": 0.0004540585938070589, + "loss": 1.8724, + "step": 2358 + }, + { + "epoch": 0.23037109375, + "grad_norm": 0.22731097042560577, + "learning_rate": 0.00045401552349840077, + "loss": 1.9341, + "step": 2359 + }, + { + "epoch": 0.23046875, + "grad_norm": 0.23395659029483795, + "learning_rate": 0.00045397243530767773, + "loss": 1.9038, + "step": 2360 + }, + { + "epoch": 0.23056640625, + "grad_norm": 0.24531114101409912, + "learning_rate": 0.000453929329239194, + "loss": 1.8877, + "step": 2361 + }, + { + "epoch": 0.2306640625, + "grad_norm": 0.25629279017448425, + "learning_rate": 0.00045388620529725546, + "loss": 1.8831, + "step": 2362 + }, + { + "epoch": 0.23076171875, + "grad_norm": 0.24789990484714508, + "learning_rate": 0.00045384306348616977, + "loss": 1.8992, + "step": 2363 + }, + { + "epoch": 0.230859375, + "grad_norm": 0.23063692450523376, + "learning_rate": 0.00045379990381024644, + "loss": 1.8645, + "step": 2364 + }, + { + "epoch": 0.23095703125, + "grad_norm": 0.274730920791626, + "learning_rate": 0.0004537567262737968, + "loss": 1.8846, + "step": 2365 + }, + { + "epoch": 0.2310546875, + "grad_norm": 0.21056944131851196, + "learning_rate": 0.00045371353088113377, + "loss": 1.8942, + "step": 2366 + }, + { + "epoch": 0.23115234375, + "grad_norm": 0.2247105836868286, + "learning_rate": 0.0004536703176365723, + "loss": 1.9012, + "step": 2367 + }, + { + "epoch": 0.23125, + "grad_norm": 0.23797035217285156, + "learning_rate": 0.00045362708654442897, + "loss": 1.8886, + "step": 2368 + }, + { + "epoch": 0.23134765625, + "grad_norm": 0.2110954225063324, + "learning_rate": 0.0004535838376090222, + "loss": 1.8937, + "step": 2369 + }, + { + "epoch": 0.2314453125, + "grad_norm": 0.2745433747768402, + "learning_rate": 0.00045354057083467217, + "loss": 1.8799, + "step": 2370 + }, + { + "epoch": 0.23154296875, + "grad_norm": 0.20481610298156738, + "learning_rate": 0.0004534972862257008, + "loss": 1.8282, + "step": 2371 + }, + { + "epoch": 0.231640625, + "grad_norm": 0.4868389368057251, + "learning_rate": 0.00045345398378643186, + "loss": 1.8975, + "step": 2372 + }, + { + "epoch": 0.23173828125, + "grad_norm": 0.2749464213848114, + "learning_rate": 0.00045341066352119096, + "loss": 1.8683, + "step": 2373 + }, + { + "epoch": 0.2318359375, + "grad_norm": 0.2716321051120758, + "learning_rate": 0.00045336732543430536, + "loss": 1.8833, + "step": 2374 + }, + { + "epoch": 0.23193359375, + "grad_norm": 0.2876909077167511, + "learning_rate": 0.00045332396953010415, + "loss": 1.881, + "step": 2375 + }, + { + "epoch": 0.23203125, + "grad_norm": 0.2515881061553955, + "learning_rate": 0.0004532805958129181, + "loss": 1.9023, + "step": 2376 + }, + { + "epoch": 0.23212890625, + "grad_norm": 0.2852384150028229, + "learning_rate": 0.00045323720428708003, + "loss": 1.924, + "step": 2377 + }, + { + "epoch": 0.2322265625, + "grad_norm": 0.22907400131225586, + "learning_rate": 0.0004531937949569243, + "loss": 1.8765, + "step": 2378 + }, + { + "epoch": 0.23232421875, + "grad_norm": 0.2648322582244873, + "learning_rate": 0.0004531503678267871, + "loss": 1.9247, + "step": 2379 + }, + { + "epoch": 0.232421875, + "grad_norm": 0.28034675121307373, + "learning_rate": 0.00045310692290100637, + "loss": 1.8948, + "step": 2380 + }, + { + "epoch": 0.23251953125, + "grad_norm": 0.29645946621894836, + "learning_rate": 0.00045306346018392197, + "loss": 1.9011, + "step": 2381 + }, + { + "epoch": 0.2326171875, + "grad_norm": 0.24943086504936218, + "learning_rate": 0.0004530199796798754, + "loss": 1.9089, + "step": 2382 + }, + { + "epoch": 0.23271484375, + "grad_norm": 0.279883474111557, + "learning_rate": 0.00045297648139321, + "loss": 1.8779, + "step": 2383 + }, + { + "epoch": 0.2328125, + "grad_norm": 0.22576992213726044, + "learning_rate": 0.00045293296532827074, + "loss": 1.9037, + "step": 2384 + }, + { + "epoch": 0.23291015625, + "grad_norm": 0.2487352192401886, + "learning_rate": 0.0004528894314894047, + "loss": 1.9089, + "step": 2385 + }, + { + "epoch": 0.2330078125, + "grad_norm": 0.25155317783355713, + "learning_rate": 0.0004528458798809603, + "loss": 1.8979, + "step": 2386 + }, + { + "epoch": 0.23310546875, + "grad_norm": 0.25363826751708984, + "learning_rate": 0.00045280231050728815, + "loss": 1.9251, + "step": 2387 + }, + { + "epoch": 0.233203125, + "grad_norm": 0.2635079026222229, + "learning_rate": 0.0004527587233727404, + "loss": 1.9152, + "step": 2388 + }, + { + "epoch": 0.23330078125, + "grad_norm": 0.2683078944683075, + "learning_rate": 0.0004527151184816709, + "loss": 1.8777, + "step": 2389 + }, + { + "epoch": 0.2333984375, + "grad_norm": 0.23932483792304993, + "learning_rate": 0.00045267149583843555, + "loss": 1.8953, + "step": 2390 + }, + { + "epoch": 0.23349609375, + "grad_norm": 0.20382820069789886, + "learning_rate": 0.00045262785544739173, + "loss": 1.9141, + "step": 2391 + }, + { + "epoch": 0.23359375, + "grad_norm": 0.21496914327144623, + "learning_rate": 0.00045258419731289884, + "loss": 1.9053, + "step": 2392 + }, + { + "epoch": 0.23369140625, + "grad_norm": 0.3160604238510132, + "learning_rate": 0.00045254052143931786, + "loss": 1.9264, + "step": 2393 + }, + { + "epoch": 0.2337890625, + "grad_norm": 0.3175326883792877, + "learning_rate": 0.0004524968278310117, + "loss": 1.8834, + "step": 2394 + }, + { + "epoch": 0.23388671875, + "grad_norm": 0.2783551514148712, + "learning_rate": 0.0004524531164923448, + "loss": 1.8688, + "step": 2395 + }, + { + "epoch": 0.233984375, + "grad_norm": 0.24283266067504883, + "learning_rate": 0.0004524093874276838, + "loss": 1.9118, + "step": 2396 + }, + { + "epoch": 0.23408203125, + "grad_norm": 0.26176488399505615, + "learning_rate": 0.0004523656406413967, + "loss": 1.9251, + "step": 2397 + }, + { + "epoch": 0.2341796875, + "grad_norm": 0.24710685014724731, + "learning_rate": 0.0004523218761378533, + "loss": 1.8714, + "step": 2398 + }, + { + "epoch": 0.23427734375, + "grad_norm": 0.21796317398548126, + "learning_rate": 0.00045227809392142546, + "loss": 1.8935, + "step": 2399 + }, + { + "epoch": 0.234375, + "grad_norm": 0.25704821944236755, + "learning_rate": 0.00045223429399648664, + "loss": 1.876, + "step": 2400 + }, + { + "epoch": 0.23447265625, + "grad_norm": 0.2729525864124298, + "learning_rate": 0.0004521904763674119, + "loss": 1.9012, + "step": 2401 + }, + { + "epoch": 0.2345703125, + "grad_norm": 0.18755966424942017, + "learning_rate": 0.00045214664103857845, + "loss": 1.8964, + "step": 2402 + }, + { + "epoch": 0.23466796875, + "grad_norm": 0.2789023816585541, + "learning_rate": 0.0004521027880143648, + "loss": 1.8949, + "step": 2403 + }, + { + "epoch": 0.234765625, + "grad_norm": 0.21166810393333435, + "learning_rate": 0.00045205891729915176, + "loss": 1.9133, + "step": 2404 + }, + { + "epoch": 0.23486328125, + "grad_norm": 0.22952231764793396, + "learning_rate": 0.00045201502889732144, + "loss": 1.9143, + "step": 2405 + }, + { + "epoch": 0.2349609375, + "grad_norm": 0.23634375631809235, + "learning_rate": 0.00045197112281325784, + "loss": 1.8645, + "step": 2406 + }, + { + "epoch": 0.23505859375, + "grad_norm": 0.25142142176628113, + "learning_rate": 0.00045192719905134705, + "loss": 1.9343, + "step": 2407 + }, + { + "epoch": 0.23515625, + "grad_norm": 0.2976498007774353, + "learning_rate": 0.0004518832576159764, + "loss": 1.918, + "step": 2408 + }, + { + "epoch": 0.23525390625, + "grad_norm": 0.2667110562324524, + "learning_rate": 0.0004518392985115354, + "loss": 1.9093, + "step": 2409 + }, + { + "epoch": 0.2353515625, + "grad_norm": 0.2608035206794739, + "learning_rate": 0.00045179532174241516, + "loss": 1.9234, + "step": 2410 + }, + { + "epoch": 0.23544921875, + "grad_norm": 0.26505768299102783, + "learning_rate": 0.0004517513273130085, + "loss": 1.8844, + "step": 2411 + }, + { + "epoch": 0.235546875, + "grad_norm": 0.31293320655822754, + "learning_rate": 0.0004517073152277101, + "loss": 1.907, + "step": 2412 + }, + { + "epoch": 0.23564453125, + "grad_norm": 0.27297648787498474, + "learning_rate": 0.00045166328549091643, + "loss": 1.9158, + "step": 2413 + }, + { + "epoch": 0.2357421875, + "grad_norm": 0.2741674780845642, + "learning_rate": 0.00045161923810702563, + "loss": 1.9011, + "step": 2414 + }, + { + "epoch": 0.23583984375, + "grad_norm": 0.29327699542045593, + "learning_rate": 0.00045157517308043774, + "loss": 1.894, + "step": 2415 + }, + { + "epoch": 0.2359375, + "grad_norm": 0.28643742203712463, + "learning_rate": 0.00045153109041555425, + "loss": 1.8993, + "step": 2416 + }, + { + "epoch": 0.23603515625, + "grad_norm": 0.46471482515335083, + "learning_rate": 0.00045148699011677873, + "loss": 1.9068, + "step": 2417 + }, + { + "epoch": 0.2361328125, + "grad_norm": 0.331559956073761, + "learning_rate": 0.0004514428721885165, + "loss": 1.9106, + "step": 2418 + }, + { + "epoch": 0.23623046875, + "grad_norm": 0.2610962688922882, + "learning_rate": 0.00045139873663517445, + "loss": 1.8757, + "step": 2419 + }, + { + "epoch": 0.236328125, + "grad_norm": 0.32770147919654846, + "learning_rate": 0.0004513545834611614, + "loss": 1.9095, + "step": 2420 + }, + { + "epoch": 0.23642578125, + "grad_norm": 0.2246219664812088, + "learning_rate": 0.00045131041267088787, + "loss": 1.8643, + "step": 2421 + }, + { + "epoch": 0.2365234375, + "grad_norm": 0.31902652978897095, + "learning_rate": 0.00045126622426876596, + "loss": 1.9061, + "step": 2422 + }, + { + "epoch": 0.23662109375, + "grad_norm": 0.20017661154270172, + "learning_rate": 0.00045122201825920983, + "loss": 1.9162, + "step": 2423 + }, + { + "epoch": 0.23671875, + "grad_norm": 0.30280932784080505, + "learning_rate": 0.0004511777946466353, + "loss": 1.8944, + "step": 2424 + }, + { + "epoch": 0.23681640625, + "grad_norm": 0.20749028027057648, + "learning_rate": 0.00045113355343545994, + "loss": 1.871, + "step": 2425 + }, + { + "epoch": 0.2369140625, + "grad_norm": 0.2936987280845642, + "learning_rate": 0.000451089294630103, + "loss": 1.8808, + "step": 2426 + }, + { + "epoch": 0.23701171875, + "grad_norm": 0.18236121535301208, + "learning_rate": 0.0004510450182349854, + "loss": 1.8534, + "step": 2427 + }, + { + "epoch": 0.237109375, + "grad_norm": 0.25331661105155945, + "learning_rate": 0.0004510007242545302, + "loss": 1.8707, + "step": 2428 + }, + { + "epoch": 0.23720703125, + "grad_norm": 0.2117713838815689, + "learning_rate": 0.00045095641269316184, + "loss": 1.9254, + "step": 2429 + }, + { + "epoch": 0.2373046875, + "grad_norm": 0.25148898363113403, + "learning_rate": 0.0004509120835553067, + "loss": 1.8973, + "step": 2430 + }, + { + "epoch": 0.23740234375, + "grad_norm": 0.24941754341125488, + "learning_rate": 0.0004508677368453929, + "loss": 1.8605, + "step": 2431 + }, + { + "epoch": 0.2375, + "grad_norm": 0.20781292021274567, + "learning_rate": 0.00045082337256785015, + "loss": 1.8802, + "step": 2432 + }, + { + "epoch": 0.23759765625, + "grad_norm": 0.2777664363384247, + "learning_rate": 0.0004507789907271102, + "loss": 1.8958, + "step": 2433 + }, + { + "epoch": 0.2376953125, + "grad_norm": 0.2164555788040161, + "learning_rate": 0.0004507345913276063, + "loss": 1.8586, + "step": 2434 + }, + { + "epoch": 0.23779296875, + "grad_norm": 0.2553316354751587, + "learning_rate": 0.0004506901743737736, + "loss": 1.8892, + "step": 2435 + }, + { + "epoch": 0.237890625, + "grad_norm": 0.23472066223621368, + "learning_rate": 0.0004506457398700489, + "loss": 1.8896, + "step": 2436 + }, + { + "epoch": 0.23798828125, + "grad_norm": 0.2276161015033722, + "learning_rate": 0.00045060128782087094, + "loss": 1.8851, + "step": 2437 + }, + { + "epoch": 0.2380859375, + "grad_norm": 0.21676012873649597, + "learning_rate": 0.00045055681823068006, + "loss": 1.8854, + "step": 2438 + }, + { + "epoch": 0.23818359375, + "grad_norm": 0.22366516292095184, + "learning_rate": 0.00045051233110391823, + "loss": 1.9309, + "step": 2439 + }, + { + "epoch": 0.23828125, + "grad_norm": 0.28527531027793884, + "learning_rate": 0.00045046782644502946, + "loss": 1.8901, + "step": 2440 + }, + { + "epoch": 0.23837890625, + "grad_norm": 0.3597792088985443, + "learning_rate": 0.0004504233042584594, + "loss": 1.8695, + "step": 2441 + }, + { + "epoch": 0.2384765625, + "grad_norm": 0.2790103554725647, + "learning_rate": 0.00045037876454865526, + "loss": 1.8756, + "step": 2442 + }, + { + "epoch": 0.23857421875, + "grad_norm": 0.3042321503162384, + "learning_rate": 0.0004503342073200663, + "loss": 1.8643, + "step": 2443 + }, + { + "epoch": 0.238671875, + "grad_norm": 0.24617038667201996, + "learning_rate": 0.0004502896325771433, + "loss": 1.9004, + "step": 2444 + }, + { + "epoch": 0.23876953125, + "grad_norm": 0.29559648036956787, + "learning_rate": 0.000450245040324339, + "loss": 1.9126, + "step": 2445 + }, + { + "epoch": 0.2388671875, + "grad_norm": 0.28401464223861694, + "learning_rate": 0.0004502004305661077, + "loss": 1.8639, + "step": 2446 + }, + { + "epoch": 0.23896484375, + "grad_norm": 0.2113400399684906, + "learning_rate": 0.00045015580330690547, + "loss": 1.8863, + "step": 2447 + }, + { + "epoch": 0.2390625, + "grad_norm": 0.28481796383857727, + "learning_rate": 0.0004501111585511903, + "loss": 1.8826, + "step": 2448 + }, + { + "epoch": 0.23916015625, + "grad_norm": 0.25684455037117004, + "learning_rate": 0.00045006649630342164, + "loss": 1.8952, + "step": 2449 + }, + { + "epoch": 0.2392578125, + "grad_norm": 0.23974579572677612, + "learning_rate": 0.00045002181656806105, + "loss": 1.8615, + "step": 2450 + }, + { + "epoch": 0.23935546875, + "grad_norm": 0.24866057932376862, + "learning_rate": 0.0004499771193495715, + "loss": 1.9343, + "step": 2451 + }, + { + "epoch": 0.239453125, + "grad_norm": 0.22083763778209686, + "learning_rate": 0.0004499324046524179, + "loss": 1.8349, + "step": 2452 + }, + { + "epoch": 0.23955078125, + "grad_norm": 0.33529961109161377, + "learning_rate": 0.00044988767248106676, + "loss": 1.9153, + "step": 2453 + }, + { + "epoch": 0.2396484375, + "grad_norm": 0.26694998145103455, + "learning_rate": 0.0004498429228399866, + "loss": 1.8993, + "step": 2454 + }, + { + "epoch": 0.23974609375, + "grad_norm": 0.29110854864120483, + "learning_rate": 0.0004497981557336474, + "loss": 1.9531, + "step": 2455 + }, + { + "epoch": 0.23984375, + "grad_norm": 0.3439907729625702, + "learning_rate": 0.00044975337116652097, + "loss": 1.8862, + "step": 2456 + }, + { + "epoch": 0.23994140625, + "grad_norm": 0.2403963953256607, + "learning_rate": 0.00044970856914308097, + "loss": 1.8965, + "step": 2457 + }, + { + "epoch": 0.2400390625, + "grad_norm": 0.28895774483680725, + "learning_rate": 0.00044966374966780266, + "loss": 1.8895, + "step": 2458 + }, + { + "epoch": 0.24013671875, + "grad_norm": 0.2802990972995758, + "learning_rate": 0.00044961891274516313, + "loss": 1.8932, + "step": 2459 + }, + { + "epoch": 0.240234375, + "grad_norm": 0.251863032579422, + "learning_rate": 0.0004495740583796413, + "loss": 1.9222, + "step": 2460 + }, + { + "epoch": 0.24033203125, + "grad_norm": 0.22680626809597015, + "learning_rate": 0.0004495291865757175, + "loss": 1.8676, + "step": 2461 + }, + { + "epoch": 0.2404296875, + "grad_norm": 0.24981464445590973, + "learning_rate": 0.0004494842973378742, + "loss": 1.8868, + "step": 2462 + }, + { + "epoch": 0.24052734375, + "grad_norm": 0.19781389832496643, + "learning_rate": 0.0004494393906705954, + "loss": 1.9079, + "step": 2463 + }, + { + "epoch": 0.240625, + "grad_norm": 0.22644692659378052, + "learning_rate": 0.00044939446657836686, + "loss": 1.9074, + "step": 2464 + }, + { + "epoch": 0.24072265625, + "grad_norm": 0.2107515037059784, + "learning_rate": 0.0004493495250656761, + "loss": 1.8994, + "step": 2465 + }, + { + "epoch": 0.2408203125, + "grad_norm": 0.18987129628658295, + "learning_rate": 0.00044930456613701236, + "loss": 1.9397, + "step": 2466 + }, + { + "epoch": 0.24091796875, + "grad_norm": 0.21435169875621796, + "learning_rate": 0.00044925958979686676, + "loss": 1.9007, + "step": 2467 + }, + { + "epoch": 0.241015625, + "grad_norm": 0.1948396861553192, + "learning_rate": 0.00044921459604973184, + "loss": 1.8762, + "step": 2468 + }, + { + "epoch": 0.24111328125, + "grad_norm": 0.2153381109237671, + "learning_rate": 0.00044916958490010217, + "loss": 1.9325, + "step": 2469 + }, + { + "epoch": 0.2412109375, + "grad_norm": 0.21237823367118835, + "learning_rate": 0.00044912455635247404, + "loss": 1.9006, + "step": 2470 + }, + { + "epoch": 0.24130859375, + "grad_norm": 0.19011497497558594, + "learning_rate": 0.0004490795104113453, + "loss": 1.9169, + "step": 2471 + }, + { + "epoch": 0.24140625, + "grad_norm": 0.2259521484375, + "learning_rate": 0.0004490344470812157, + "loss": 1.8984, + "step": 2472 + }, + { + "epoch": 0.24150390625, + "grad_norm": 0.1522991955280304, + "learning_rate": 0.0004489893663665866, + "loss": 1.838, + "step": 2473 + }, + { + "epoch": 0.2416015625, + "grad_norm": 0.21235786378383636, + "learning_rate": 0.00044894426827196124, + "loss": 1.9033, + "step": 2474 + }, + { + "epoch": 0.24169921875, + "grad_norm": 0.20965880155563354, + "learning_rate": 0.0004488991528018445, + "loss": 1.9381, + "step": 2475 + }, + { + "epoch": 0.241796875, + "grad_norm": 0.25868213176727295, + "learning_rate": 0.0004488540199607429, + "loss": 1.8907, + "step": 2476 + }, + { + "epoch": 0.24189453125, + "grad_norm": 0.2545277178287506, + "learning_rate": 0.00044880886975316503, + "loss": 1.8905, + "step": 2477 + }, + { + "epoch": 0.2419921875, + "grad_norm": 0.17712853848934174, + "learning_rate": 0.00044876370218362085, + "loss": 1.9022, + "step": 2478 + }, + { + "epoch": 0.24208984375, + "grad_norm": 0.24900510907173157, + "learning_rate": 0.0004487185172566222, + "loss": 1.9247, + "step": 2479 + }, + { + "epoch": 0.2421875, + "grad_norm": 0.24943552911281586, + "learning_rate": 0.0004486733149766827, + "loss": 1.9378, + "step": 2480 + }, + { + "epoch": 0.24228515625, + "grad_norm": 0.27004435658454895, + "learning_rate": 0.0004486280953483177, + "loss": 1.8644, + "step": 2481 + }, + { + "epoch": 0.2423828125, + "grad_norm": 0.21977658569812775, + "learning_rate": 0.0004485828583760441, + "loss": 1.9166, + "step": 2482 + }, + { + "epoch": 0.24248046875, + "grad_norm": 0.17602773010730743, + "learning_rate": 0.0004485376040643808, + "loss": 1.8733, + "step": 2483 + }, + { + "epoch": 0.242578125, + "grad_norm": 0.2778310179710388, + "learning_rate": 0.0004484923324178483, + "loss": 1.8788, + "step": 2484 + }, + { + "epoch": 0.24267578125, + "grad_norm": 0.33434948325157166, + "learning_rate": 0.00044844704344096863, + "loss": 1.8932, + "step": 2485 + }, + { + "epoch": 0.2427734375, + "grad_norm": 0.2541770040988922, + "learning_rate": 0.00044840173713826603, + "loss": 1.8727, + "step": 2486 + }, + { + "epoch": 0.24287109375, + "grad_norm": 0.21049551665782928, + "learning_rate": 0.00044835641351426614, + "loss": 1.8567, + "step": 2487 + }, + { + "epoch": 0.24296875, + "grad_norm": 0.2456684708595276, + "learning_rate": 0.0004483110725734962, + "loss": 1.907, + "step": 2488 + }, + { + "epoch": 0.24306640625, + "grad_norm": 0.2708995044231415, + "learning_rate": 0.0004482657143204857, + "loss": 1.9146, + "step": 2489 + }, + { + "epoch": 0.2431640625, + "grad_norm": 0.2767466604709625, + "learning_rate": 0.00044822033875976513, + "loss": 1.9058, + "step": 2490 + }, + { + "epoch": 0.24326171875, + "grad_norm": 0.21536017954349518, + "learning_rate": 0.00044817494589586746, + "loss": 1.8867, + "step": 2491 + }, + { + "epoch": 0.243359375, + "grad_norm": 0.32470428943634033, + "learning_rate": 0.0004481295357333268, + "loss": 1.8651, + "step": 2492 + }, + { + "epoch": 0.24345703125, + "grad_norm": 0.35625115036964417, + "learning_rate": 0.00044808410827667936, + "loss": 1.8959, + "step": 2493 + }, + { + "epoch": 0.2435546875, + "grad_norm": 0.2434857189655304, + "learning_rate": 0.0004480386635304629, + "loss": 1.8805, + "step": 2494 + }, + { + "epoch": 0.24365234375, + "grad_norm": 0.2814987003803253, + "learning_rate": 0.000447993201499217, + "loss": 1.9388, + "step": 2495 + }, + { + "epoch": 0.24375, + "grad_norm": 0.3357986807823181, + "learning_rate": 0.00044794772218748275, + "loss": 1.9182, + "step": 2496 + }, + { + "epoch": 0.24384765625, + "grad_norm": 0.26190170645713806, + "learning_rate": 0.00044790222559980313, + "loss": 1.9124, + "step": 2497 + }, + { + "epoch": 0.2439453125, + "grad_norm": 0.2307833731174469, + "learning_rate": 0.0004478567117407231, + "loss": 1.9049, + "step": 2498 + }, + { + "epoch": 0.24404296875, + "grad_norm": 0.24210835993289948, + "learning_rate": 0.00044781118061478897, + "loss": 1.8924, + "step": 2499 + }, + { + "epoch": 0.244140625, + "grad_norm": 0.2340785562992096, + "learning_rate": 0.00044776563222654876, + "loss": 1.8846, + "step": 2500 + }, + { + "epoch": 0.24423828125, + "grad_norm": 0.20149527490139008, + "learning_rate": 0.0004477200665805525, + "loss": 1.8752, + "step": 2501 + }, + { + "epoch": 0.2443359375, + "grad_norm": 0.21381939947605133, + "learning_rate": 0.00044767448368135176, + "loss": 1.8537, + "step": 2502 + }, + { + "epoch": 0.24443359375, + "grad_norm": 0.20310865342617035, + "learning_rate": 0.0004476288835334999, + "loss": 1.926, + "step": 2503 + }, + { + "epoch": 0.24453125, + "grad_norm": 0.21173352003097534, + "learning_rate": 0.0004475832661415518, + "loss": 1.9143, + "step": 2504 + }, + { + "epoch": 0.24462890625, + "grad_norm": 0.22699382901191711, + "learning_rate": 0.0004475376315100645, + "loss": 1.9025, + "step": 2505 + }, + { + "epoch": 0.2447265625, + "grad_norm": 0.2502591907978058, + "learning_rate": 0.0004474919796435963, + "loss": 1.9133, + "step": 2506 + }, + { + "epoch": 0.24482421875, + "grad_norm": 0.18316835165023804, + "learning_rate": 0.00044744631054670743, + "loss": 1.8408, + "step": 2507 + }, + { + "epoch": 0.244921875, + "grad_norm": 0.2320968210697174, + "learning_rate": 0.00044740062422395994, + "loss": 1.924, + "step": 2508 + }, + { + "epoch": 0.24501953125, + "grad_norm": 0.22780783474445343, + "learning_rate": 0.0004473549206799174, + "loss": 1.8782, + "step": 2509 + }, + { + "epoch": 0.2451171875, + "grad_norm": 0.22676895558834076, + "learning_rate": 0.0004473091999191452, + "loss": 1.8904, + "step": 2510 + }, + { + "epoch": 0.24521484375, + "grad_norm": 0.22365321218967438, + "learning_rate": 0.0004472634619462105, + "loss": 1.9029, + "step": 2511 + }, + { + "epoch": 0.2453125, + "grad_norm": 0.24881736934185028, + "learning_rate": 0.00044721770676568203, + "loss": 1.8722, + "step": 2512 + }, + { + "epoch": 0.24541015625, + "grad_norm": 0.1900773048400879, + "learning_rate": 0.0004471719343821303, + "loss": 1.9069, + "step": 2513 + }, + { + "epoch": 0.2455078125, + "grad_norm": 0.253597229719162, + "learning_rate": 0.00044712614480012773, + "loss": 1.9053, + "step": 2514 + }, + { + "epoch": 0.24560546875, + "grad_norm": 0.2389373481273651, + "learning_rate": 0.0004470803380242481, + "loss": 1.892, + "step": 2515 + }, + { + "epoch": 0.245703125, + "grad_norm": 0.21958309412002563, + "learning_rate": 0.0004470345140590673, + "loss": 1.9074, + "step": 2516 + }, + { + "epoch": 0.24580078125, + "grad_norm": 0.24225091934204102, + "learning_rate": 0.00044698867290916263, + "loss": 1.8854, + "step": 2517 + }, + { + "epoch": 0.2458984375, + "grad_norm": 0.19851315021514893, + "learning_rate": 0.0004469428145791132, + "loss": 1.9327, + "step": 2518 + }, + { + "epoch": 0.24599609375, + "grad_norm": 0.27460166811943054, + "learning_rate": 0.0004468969390734998, + "loss": 1.8819, + "step": 2519 + }, + { + "epoch": 0.24609375, + "grad_norm": 0.250698447227478, + "learning_rate": 0.00044685104639690515, + "loss": 1.8329, + "step": 2520 + }, + { + "epoch": 0.24619140625, + "grad_norm": 0.22741656005382538, + "learning_rate": 0.0004468051365539133, + "loss": 1.9376, + "step": 2521 + }, + { + "epoch": 0.2462890625, + "grad_norm": 0.2052220106124878, + "learning_rate": 0.00044675920954911045, + "loss": 1.9158, + "step": 2522 + }, + { + "epoch": 0.24638671875, + "grad_norm": 0.20518803596496582, + "learning_rate": 0.0004467132653870842, + "loss": 1.8723, + "step": 2523 + }, + { + "epoch": 0.246484375, + "grad_norm": 0.2157951146364212, + "learning_rate": 0.00044666730407242407, + "loss": 1.8463, + "step": 2524 + }, + { + "epoch": 0.24658203125, + "grad_norm": 0.2367670238018036, + "learning_rate": 0.000446621325609721, + "loss": 1.9047, + "step": 2525 + }, + { + "epoch": 0.2466796875, + "grad_norm": 0.2216240018606186, + "learning_rate": 0.00044657533000356793, + "loss": 1.8725, + "step": 2526 + }, + { + "epoch": 0.24677734375, + "grad_norm": 0.20380742847919464, + "learning_rate": 0.00044652931725855946, + "loss": 1.9393, + "step": 2527 + }, + { + "epoch": 0.246875, + "grad_norm": 0.2439001351594925, + "learning_rate": 0.00044648328737929176, + "loss": 1.8832, + "step": 2528 + }, + { + "epoch": 0.24697265625, + "grad_norm": 0.19296857714653015, + "learning_rate": 0.00044643724037036293, + "loss": 1.8894, + "step": 2529 + }, + { + "epoch": 0.2470703125, + "grad_norm": 0.23245202004909515, + "learning_rate": 0.00044639117623637253, + "loss": 1.9035, + "step": 2530 + }, + { + "epoch": 0.24716796875, + "grad_norm": 0.23975121974945068, + "learning_rate": 0.00044634509498192197, + "loss": 1.9336, + "step": 2531 + }, + { + "epoch": 0.247265625, + "grad_norm": 0.20720966160297394, + "learning_rate": 0.0004462989966116145, + "loss": 1.9538, + "step": 2532 + }, + { + "epoch": 0.24736328125, + "grad_norm": 0.23715616762638092, + "learning_rate": 0.0004462528811300548, + "loss": 1.8944, + "step": 2533 + }, + { + "epoch": 0.2474609375, + "grad_norm": 0.22838465869426727, + "learning_rate": 0.00044620674854184937, + "loss": 1.8472, + "step": 2534 + }, + { + "epoch": 0.24755859375, + "grad_norm": 0.19445407390594482, + "learning_rate": 0.00044616059885160657, + "loss": 1.9021, + "step": 2535 + }, + { + "epoch": 0.24765625, + "grad_norm": 0.22580035030841827, + "learning_rate": 0.00044611443206393634, + "loss": 1.9035, + "step": 2536 + }, + { + "epoch": 0.24775390625, + "grad_norm": 0.18850651383399963, + "learning_rate": 0.0004460682481834503, + "loss": 1.8769, + "step": 2537 + }, + { + "epoch": 0.2478515625, + "grad_norm": 0.20952773094177246, + "learning_rate": 0.0004460220472147617, + "loss": 1.9035, + "step": 2538 + }, + { + "epoch": 0.24794921875, + "grad_norm": 0.2210458666086197, + "learning_rate": 0.00044597582916248567, + "loss": 1.882, + "step": 2539 + }, + { + "epoch": 0.248046875, + "grad_norm": 0.22584138810634613, + "learning_rate": 0.0004459295940312391, + "loss": 1.9031, + "step": 2540 + }, + { + "epoch": 0.24814453125, + "grad_norm": 0.22548091411590576, + "learning_rate": 0.0004458833418256404, + "loss": 1.9215, + "step": 2541 + }, + { + "epoch": 0.2482421875, + "grad_norm": 0.22244450449943542, + "learning_rate": 0.0004458370725503097, + "loss": 1.8845, + "step": 2542 + }, + { + "epoch": 0.24833984375, + "grad_norm": 0.25378334522247314, + "learning_rate": 0.00044579078620986896, + "loss": 1.908, + "step": 2543 + }, + { + "epoch": 0.2484375, + "grad_norm": 0.3173237144947052, + "learning_rate": 0.0004457444828089417, + "loss": 1.9347, + "step": 2544 + }, + { + "epoch": 0.24853515625, + "grad_norm": 0.3610992431640625, + "learning_rate": 0.00044569816235215333, + "loss": 1.8956, + "step": 2545 + }, + { + "epoch": 0.2486328125, + "grad_norm": 0.24675659835338593, + "learning_rate": 0.00044565182484413075, + "loss": 1.905, + "step": 2546 + }, + { + "epoch": 0.24873046875, + "grad_norm": 0.22318989038467407, + "learning_rate": 0.0004456054702895027, + "loss": 1.8726, + "step": 2547 + }, + { + "epoch": 0.248828125, + "grad_norm": 0.3162762522697449, + "learning_rate": 0.0004455590986928996, + "loss": 1.8969, + "step": 2548 + }, + { + "epoch": 0.24892578125, + "grad_norm": 0.2868858575820923, + "learning_rate": 0.0004455127100589536, + "loss": 1.8387, + "step": 2549 + }, + { + "epoch": 0.2490234375, + "grad_norm": 0.20986978709697723, + "learning_rate": 0.0004454663043922984, + "loss": 1.86, + "step": 2550 + }, + { + "epoch": 0.24912109375, + "grad_norm": 0.20684708654880524, + "learning_rate": 0.0004454198816975697, + "loss": 1.8546, + "step": 2551 + }, + { + "epoch": 0.24921875, + "grad_norm": 0.2441062480211258, + "learning_rate": 0.00044537344197940457, + "loss": 1.8749, + "step": 2552 + }, + { + "epoch": 0.24931640625, + "grad_norm": 0.2610075771808624, + "learning_rate": 0.00044532698524244184, + "loss": 1.8743, + "step": 2553 + }, + { + "epoch": 0.2494140625, + "grad_norm": 0.21725575625896454, + "learning_rate": 0.0004452805114913223, + "loss": 1.942, + "step": 2554 + }, + { + "epoch": 0.24951171875, + "grad_norm": 0.3064901828765869, + "learning_rate": 0.0004452340207306883, + "loss": 1.8555, + "step": 2555 + }, + { + "epoch": 0.249609375, + "grad_norm": 0.3561396300792694, + "learning_rate": 0.00044518751296518367, + "loss": 1.9103, + "step": 2556 + }, + { + "epoch": 0.24970703125, + "grad_norm": 0.22716249525547028, + "learning_rate": 0.0004451409881994543, + "loss": 1.8788, + "step": 2557 + }, + { + "epoch": 0.2498046875, + "grad_norm": 0.20097126066684723, + "learning_rate": 0.00044509444643814745, + "loss": 1.8081, + "step": 2558 + }, + { + "epoch": 0.24990234375, + "grad_norm": 0.2304898500442505, + "learning_rate": 0.0004450478876859123, + "loss": 1.8918, + "step": 2559 + }, + { + "epoch": 0.25, + "grad_norm": 0.22153149545192719, + "learning_rate": 0.00044500131194739976, + "loss": 1.8949, + "step": 2560 + }, + { + "epoch": 0.25009765625, + "grad_norm": 0.21728168427944183, + "learning_rate": 0.0004449547192272622, + "loss": 1.8991, + "step": 2561 + }, + { + "epoch": 0.2501953125, + "grad_norm": 0.191215381026268, + "learning_rate": 0.00044490810953015376, + "loss": 1.8753, + "step": 2562 + }, + { + "epoch": 0.25029296875, + "grad_norm": 0.20973442494869232, + "learning_rate": 0.00044486148286073047, + "loss": 1.8833, + "step": 2563 + }, + { + "epoch": 0.250390625, + "grad_norm": 0.16450902819633484, + "learning_rate": 0.00044481483922364986, + "loss": 1.8508, + "step": 2564 + }, + { + "epoch": 0.25048828125, + "grad_norm": 0.19612371921539307, + "learning_rate": 0.00044476817862357135, + "loss": 1.8916, + "step": 2565 + }, + { + "epoch": 0.2505859375, + "grad_norm": 0.17144112288951874, + "learning_rate": 0.00044472150106515565, + "loss": 1.9255, + "step": 2566 + }, + { + "epoch": 0.25068359375, + "grad_norm": 0.2133835256099701, + "learning_rate": 0.00044467480655306565, + "loss": 1.8678, + "step": 2567 + }, + { + "epoch": 0.25078125, + "grad_norm": 0.21594852209091187, + "learning_rate": 0.0004446280950919657, + "loss": 1.8686, + "step": 2568 + }, + { + "epoch": 0.25087890625, + "grad_norm": 0.24870210886001587, + "learning_rate": 0.00044458136668652185, + "loss": 1.9342, + "step": 2569 + }, + { + "epoch": 0.2509765625, + "grad_norm": 0.2506871223449707, + "learning_rate": 0.0004445346213414017, + "loss": 1.9142, + "step": 2570 + }, + { + "epoch": 0.25107421875, + "grad_norm": 0.21057821810245514, + "learning_rate": 0.0004444878590612749, + "loss": 1.8809, + "step": 2571 + }, + { + "epoch": 0.251171875, + "grad_norm": 0.2578639090061188, + "learning_rate": 0.0004444410798508125, + "loss": 1.8706, + "step": 2572 + }, + { + "epoch": 0.25126953125, + "grad_norm": 0.23633147776126862, + "learning_rate": 0.00044439428371468724, + "loss": 1.8912, + "step": 2573 + }, + { + "epoch": 0.2513671875, + "grad_norm": 0.222493976354599, + "learning_rate": 0.00044434747065757383, + "loss": 1.8508, + "step": 2574 + }, + { + "epoch": 0.25146484375, + "grad_norm": 0.2404196411371231, + "learning_rate": 0.00044430064068414843, + "loss": 1.8879, + "step": 2575 + }, + { + "epoch": 0.2515625, + "grad_norm": 0.2464357167482376, + "learning_rate": 0.0004442537937990889, + "loss": 1.9289, + "step": 2576 + }, + { + "epoch": 0.25166015625, + "grad_norm": 0.2768436372280121, + "learning_rate": 0.0004442069300070747, + "loss": 1.8553, + "step": 2577 + }, + { + "epoch": 0.2517578125, + "grad_norm": 0.22688689827919006, + "learning_rate": 0.0004441600493127873, + "loss": 1.8665, + "step": 2578 + }, + { + "epoch": 0.25185546875, + "grad_norm": 0.28142839670181274, + "learning_rate": 0.0004441131517209096, + "loss": 1.8982, + "step": 2579 + }, + { + "epoch": 0.251953125, + "grad_norm": 0.22705985605716705, + "learning_rate": 0.0004440662372361262, + "loss": 1.8778, + "step": 2580 + }, + { + "epoch": 0.25205078125, + "grad_norm": 0.24169547855854034, + "learning_rate": 0.0004440193058631236, + "loss": 1.9192, + "step": 2581 + }, + { + "epoch": 0.2521484375, + "grad_norm": 0.2988471984863281, + "learning_rate": 0.00044397235760658963, + "loss": 1.8754, + "step": 2582 + }, + { + "epoch": 0.25224609375, + "grad_norm": 0.343605101108551, + "learning_rate": 0.0004439253924712141, + "loss": 1.8933, + "step": 2583 + }, + { + "epoch": 0.25234375, + "grad_norm": 0.3238380551338196, + "learning_rate": 0.00044387841046168844, + "loss": 1.8798, + "step": 2584 + }, + { + "epoch": 0.25244140625, + "grad_norm": 0.21151965856552124, + "learning_rate": 0.0004438314115827057, + "loss": 1.88, + "step": 2585 + }, + { + "epoch": 0.2525390625, + "grad_norm": 0.3310568630695343, + "learning_rate": 0.00044378439583896064, + "loss": 1.9042, + "step": 2586 + }, + { + "epoch": 0.25263671875, + "grad_norm": 0.30944469571113586, + "learning_rate": 0.00044373736323514973, + "loss": 1.8837, + "step": 2587 + }, + { + "epoch": 0.252734375, + "grad_norm": 0.23433548212051392, + "learning_rate": 0.00044369031377597114, + "loss": 1.9025, + "step": 2588 + }, + { + "epoch": 0.25283203125, + "grad_norm": 0.2772672176361084, + "learning_rate": 0.0004436432474661246, + "loss": 1.8573, + "step": 2589 + }, + { + "epoch": 0.2529296875, + "grad_norm": 0.251200407743454, + "learning_rate": 0.00044359616431031176, + "loss": 1.869, + "step": 2590 + }, + { + "epoch": 0.25302734375, + "grad_norm": 0.28617560863494873, + "learning_rate": 0.0004435490643132357, + "loss": 1.91, + "step": 2591 + }, + { + "epoch": 0.253125, + "grad_norm": 0.23629052937030792, + "learning_rate": 0.0004435019474796013, + "loss": 1.9149, + "step": 2592 + }, + { + "epoch": 0.25322265625, + "grad_norm": 0.2794038653373718, + "learning_rate": 0.0004434548138141152, + "loss": 1.9415, + "step": 2593 + }, + { + "epoch": 0.2533203125, + "grad_norm": 0.22255775332450867, + "learning_rate": 0.0004434076633214855, + "loss": 1.8657, + "step": 2594 + }, + { + "epoch": 0.25341796875, + "grad_norm": 0.28047165274620056, + "learning_rate": 0.00044336049600642225, + "loss": 1.9193, + "step": 2595 + }, + { + "epoch": 0.253515625, + "grad_norm": 0.16849219799041748, + "learning_rate": 0.0004433133118736369, + "loss": 1.8809, + "step": 2596 + }, + { + "epoch": 0.25361328125, + "grad_norm": 0.26295021176338196, + "learning_rate": 0.00044326611092784297, + "loss": 1.8732, + "step": 2597 + }, + { + "epoch": 0.2537109375, + "grad_norm": 0.2265648990869522, + "learning_rate": 0.00044321889317375517, + "loss": 1.8861, + "step": 2598 + }, + { + "epoch": 0.25380859375, + "grad_norm": 0.28177958726882935, + "learning_rate": 0.0004431716586160902, + "loss": 1.9275, + "step": 2599 + }, + { + "epoch": 0.25390625, + "grad_norm": 0.34475746750831604, + "learning_rate": 0.0004431244072595665, + "loss": 1.8884, + "step": 2600 + }, + { + "epoch": 0.25400390625, + "grad_norm": 0.32186293601989746, + "learning_rate": 0.0004430771391089038, + "loss": 1.9329, + "step": 2601 + }, + { + "epoch": 0.2541015625, + "grad_norm": 0.2824218273162842, + "learning_rate": 0.00044302985416882405, + "loss": 1.9011, + "step": 2602 + }, + { + "epoch": 0.25419921875, + "grad_norm": 0.33732739090919495, + "learning_rate": 0.0004429825524440505, + "loss": 1.9148, + "step": 2603 + }, + { + "epoch": 0.254296875, + "grad_norm": 0.23914110660552979, + "learning_rate": 0.00044293523393930807, + "loss": 1.866, + "step": 2604 + }, + { + "epoch": 0.25439453125, + "grad_norm": 0.2791132926940918, + "learning_rate": 0.0004428878986593236, + "loss": 1.8849, + "step": 2605 + }, + { + "epoch": 0.2544921875, + "grad_norm": 0.20789825916290283, + "learning_rate": 0.0004428405466088253, + "loss": 1.9007, + "step": 2606 + }, + { + "epoch": 0.25458984375, + "grad_norm": 0.27687475085258484, + "learning_rate": 0.0004427931777925435, + "loss": 1.8936, + "step": 2607 + }, + { + "epoch": 0.2546875, + "grad_norm": 0.29951584339141846, + "learning_rate": 0.00044274579221520964, + "loss": 1.8735, + "step": 2608 + }, + { + "epoch": 0.25478515625, + "grad_norm": 0.25301870703697205, + "learning_rate": 0.00044269838988155726, + "loss": 1.8829, + "step": 2609 + }, + { + "epoch": 0.2548828125, + "grad_norm": 0.254138320684433, + "learning_rate": 0.0004426509707963214, + "loss": 1.8881, + "step": 2610 + }, + { + "epoch": 0.25498046875, + "grad_norm": 0.2853766679763794, + "learning_rate": 0.00044260353496423883, + "loss": 1.9376, + "step": 2611 + }, + { + "epoch": 0.255078125, + "grad_norm": 0.22820377349853516, + "learning_rate": 0.00044255608239004795, + "loss": 1.8961, + "step": 2612 + }, + { + "epoch": 0.25517578125, + "grad_norm": 0.269482284784317, + "learning_rate": 0.00044250861307848884, + "loss": 1.8892, + "step": 2613 + }, + { + "epoch": 0.2552734375, + "grad_norm": 0.21227295696735382, + "learning_rate": 0.0004424611270343033, + "loss": 1.9393, + "step": 2614 + }, + { + "epoch": 0.25537109375, + "grad_norm": 0.21992765367031097, + "learning_rate": 0.00044241362426223463, + "loss": 1.8699, + "step": 2615 + }, + { + "epoch": 0.25546875, + "grad_norm": 0.26295095682144165, + "learning_rate": 0.0004423661047670282, + "loss": 1.9119, + "step": 2616 + }, + { + "epoch": 0.25556640625, + "grad_norm": 0.1936780959367752, + "learning_rate": 0.00044231856855343055, + "loss": 1.8939, + "step": 2617 + }, + { + "epoch": 0.2556640625, + "grad_norm": 0.27874645590782166, + "learning_rate": 0.0004422710156261903, + "loss": 1.9553, + "step": 2618 + }, + { + "epoch": 0.25576171875, + "grad_norm": 0.2704470753669739, + "learning_rate": 0.0004422234459900574, + "loss": 1.8742, + "step": 2619 + }, + { + "epoch": 0.255859375, + "grad_norm": 0.29514744877815247, + "learning_rate": 0.00044217585964978365, + "loss": 1.8803, + "step": 2620 + }, + { + "epoch": 0.25595703125, + "grad_norm": 0.22899918258190155, + "learning_rate": 0.0004421282566101226, + "loss": 1.8898, + "step": 2621 + }, + { + "epoch": 0.2560546875, + "grad_norm": 0.3023613691329956, + "learning_rate": 0.00044208063687582944, + "loss": 1.917, + "step": 2622 + }, + { + "epoch": 0.25615234375, + "grad_norm": 0.2797189950942993, + "learning_rate": 0.0004420330004516608, + "loss": 1.89, + "step": 2623 + }, + { + "epoch": 0.25625, + "grad_norm": 0.23178371787071228, + "learning_rate": 0.0004419853473423751, + "loss": 1.8528, + "step": 2624 + }, + { + "epoch": 0.25634765625, + "grad_norm": 0.2943453788757324, + "learning_rate": 0.0004419376775527326, + "loss": 1.8675, + "step": 2625 + }, + { + "epoch": 0.2564453125, + "grad_norm": 0.2193489670753479, + "learning_rate": 0.00044188999108749516, + "loss": 1.8843, + "step": 2626 + }, + { + "epoch": 0.25654296875, + "grad_norm": 0.2941931486129761, + "learning_rate": 0.00044184228795142607, + "loss": 1.8976, + "step": 2627 + }, + { + "epoch": 0.256640625, + "grad_norm": 0.22620061039924622, + "learning_rate": 0.00044179456814929044, + "loss": 1.8674, + "step": 2628 + }, + { + "epoch": 0.25673828125, + "grad_norm": 0.2290557473897934, + "learning_rate": 0.0004417468316858552, + "loss": 1.9054, + "step": 2629 + }, + { + "epoch": 0.2568359375, + "grad_norm": 0.24180883169174194, + "learning_rate": 0.00044169907856588865, + "loss": 1.8484, + "step": 2630 + }, + { + "epoch": 0.25693359375, + "grad_norm": 0.19765223562717438, + "learning_rate": 0.00044165130879416097, + "loss": 1.8613, + "step": 2631 + }, + { + "epoch": 0.25703125, + "grad_norm": 0.20855452120304108, + "learning_rate": 0.00044160352237544406, + "loss": 1.8621, + "step": 2632 + }, + { + "epoch": 0.25712890625, + "grad_norm": 0.23405295610427856, + "learning_rate": 0.00044155571931451105, + "loss": 1.8958, + "step": 2633 + }, + { + "epoch": 0.2572265625, + "grad_norm": 0.27464863657951355, + "learning_rate": 0.0004415078996161374, + "loss": 1.8959, + "step": 2634 + }, + { + "epoch": 0.25732421875, + "grad_norm": 0.2642795145511627, + "learning_rate": 0.00044146006328509956, + "loss": 1.8968, + "step": 2635 + }, + { + "epoch": 0.257421875, + "grad_norm": 0.29816219210624695, + "learning_rate": 0.00044141221032617626, + "loss": 1.8845, + "step": 2636 + }, + { + "epoch": 0.25751953125, + "grad_norm": 0.251708984375, + "learning_rate": 0.0004413643407441473, + "loss": 1.9025, + "step": 2637 + }, + { + "epoch": 0.2576171875, + "grad_norm": 0.28949257731437683, + "learning_rate": 0.0004413164545437946, + "loss": 1.8946, + "step": 2638 + }, + { + "epoch": 0.25771484375, + "grad_norm": 0.24704940617084503, + "learning_rate": 0.0004412685517299015, + "loss": 1.8745, + "step": 2639 + }, + { + "epoch": 0.2578125, + "grad_norm": 0.25325044989585876, + "learning_rate": 0.0004412206323072532, + "loss": 1.8796, + "step": 2640 + }, + { + "epoch": 0.25791015625, + "grad_norm": 0.22802412509918213, + "learning_rate": 0.00044117269628063616, + "loss": 1.909, + "step": 2641 + }, + { + "epoch": 0.2580078125, + "grad_norm": 0.21644683182239532, + "learning_rate": 0.00044112474365483903, + "loss": 1.846, + "step": 2642 + }, + { + "epoch": 0.25810546875, + "grad_norm": 0.20541459321975708, + "learning_rate": 0.00044107677443465165, + "loss": 1.8854, + "step": 2643 + }, + { + "epoch": 0.258203125, + "grad_norm": 0.2531701326370239, + "learning_rate": 0.00044102878862486587, + "loss": 1.9159, + "step": 2644 + }, + { + "epoch": 0.25830078125, + "grad_norm": 0.20743289589881897, + "learning_rate": 0.00044098078623027495, + "loss": 1.8419, + "step": 2645 + }, + { + "epoch": 0.2583984375, + "grad_norm": 0.23902501165866852, + "learning_rate": 0.00044093276725567403, + "loss": 1.883, + "step": 2646 + }, + { + "epoch": 0.25849609375, + "grad_norm": 0.260051965713501, + "learning_rate": 0.00044088473170585964, + "loss": 1.9066, + "step": 2647 + }, + { + "epoch": 0.25859375, + "grad_norm": 0.22198320925235748, + "learning_rate": 0.00044083667958563007, + "loss": 1.8814, + "step": 2648 + }, + { + "epoch": 0.25869140625, + "grad_norm": 0.20246773958206177, + "learning_rate": 0.0004407886108997855, + "loss": 1.923, + "step": 2649 + }, + { + "epoch": 0.2587890625, + "grad_norm": 0.26774612069129944, + "learning_rate": 0.0004407405256531274, + "loss": 1.8811, + "step": 2650 + }, + { + "epoch": 0.25888671875, + "grad_norm": 0.2821747064590454, + "learning_rate": 0.0004406924238504592, + "loss": 1.9521, + "step": 2651 + }, + { + "epoch": 0.258984375, + "grad_norm": 0.2747217118740082, + "learning_rate": 0.0004406443054965857, + "loss": 1.863, + "step": 2652 + }, + { + "epoch": 0.25908203125, + "grad_norm": 0.24007029831409454, + "learning_rate": 0.0004405961705963135, + "loss": 1.8777, + "step": 2653 + }, + { + "epoch": 0.2591796875, + "grad_norm": 0.2279479056596756, + "learning_rate": 0.000440548019154451, + "loss": 1.8902, + "step": 2654 + }, + { + "epoch": 0.25927734375, + "grad_norm": 0.24569466710090637, + "learning_rate": 0.00044049985117580796, + "loss": 1.8834, + "step": 2655 + }, + { + "epoch": 0.259375, + "grad_norm": 0.22550983726978302, + "learning_rate": 0.00044045166666519604, + "loss": 1.8953, + "step": 2656 + }, + { + "epoch": 0.25947265625, + "grad_norm": 0.2546038031578064, + "learning_rate": 0.00044040346562742834, + "loss": 1.8834, + "step": 2657 + }, + { + "epoch": 0.2595703125, + "grad_norm": 0.21834610402584076, + "learning_rate": 0.00044035524806731975, + "loss": 1.8543, + "step": 2658 + }, + { + "epoch": 0.25966796875, + "grad_norm": 0.22710272669792175, + "learning_rate": 0.0004403070139896869, + "loss": 1.8995, + "step": 2659 + }, + { + "epoch": 0.259765625, + "grad_norm": 0.19748175144195557, + "learning_rate": 0.0004402587633993477, + "loss": 1.864, + "step": 2660 + }, + { + "epoch": 0.25986328125, + "grad_norm": 0.23321975767612457, + "learning_rate": 0.0004402104963011222, + "loss": 1.9256, + "step": 2661 + }, + { + "epoch": 0.2599609375, + "grad_norm": 0.19426730275154114, + "learning_rate": 0.0004401622126998317, + "loss": 1.8956, + "step": 2662 + }, + { + "epoch": 0.26005859375, + "grad_norm": 0.2513621747493744, + "learning_rate": 0.0004401139126002993, + "loss": 1.9298, + "step": 2663 + }, + { + "epoch": 0.26015625, + "grad_norm": 0.2774468660354614, + "learning_rate": 0.00044006559600734993, + "loss": 1.8946, + "step": 2664 + }, + { + "epoch": 0.26025390625, + "grad_norm": 0.2116554081439972, + "learning_rate": 0.00044001726292580977, + "loss": 1.8713, + "step": 2665 + }, + { + "epoch": 0.2603515625, + "grad_norm": 0.24314185976982117, + "learning_rate": 0.000439968913360507, + "loss": 1.8807, + "step": 2666 + }, + { + "epoch": 0.26044921875, + "grad_norm": 0.22041741013526917, + "learning_rate": 0.00043992054731627135, + "loss": 1.9022, + "step": 2667 + }, + { + "epoch": 0.260546875, + "grad_norm": 0.214686781167984, + "learning_rate": 0.00043987216479793404, + "loss": 1.8687, + "step": 2668 + }, + { + "epoch": 0.26064453125, + "grad_norm": 0.254207581281662, + "learning_rate": 0.0004398237658103281, + "loss": 1.8489, + "step": 2669 + }, + { + "epoch": 0.2607421875, + "grad_norm": 0.21913522481918335, + "learning_rate": 0.0004397753503582881, + "loss": 1.9079, + "step": 2670 + }, + { + "epoch": 0.26083984375, + "grad_norm": 0.2150515615940094, + "learning_rate": 0.0004397269184466505, + "loss": 1.8985, + "step": 2671 + }, + { + "epoch": 0.2609375, + "grad_norm": 0.26243162155151367, + "learning_rate": 0.000439678470080253, + "loss": 1.8512, + "step": 2672 + }, + { + "epoch": 0.26103515625, + "grad_norm": 0.20125195384025574, + "learning_rate": 0.0004396300052639353, + "loss": 1.8621, + "step": 2673 + }, + { + "epoch": 0.2611328125, + "grad_norm": 0.2900947630405426, + "learning_rate": 0.00043958152400253855, + "loss": 1.9354, + "step": 2674 + }, + { + "epoch": 0.26123046875, + "grad_norm": 0.2357524037361145, + "learning_rate": 0.00043953302630090566, + "loss": 1.9136, + "step": 2675 + }, + { + "epoch": 0.261328125, + "grad_norm": 0.23963971436023712, + "learning_rate": 0.000439484512163881, + "loss": 1.867, + "step": 2676 + }, + { + "epoch": 0.26142578125, + "grad_norm": 0.27860602736473083, + "learning_rate": 0.00043943598159631075, + "loss": 1.9149, + "step": 2677 + }, + { + "epoch": 0.2615234375, + "grad_norm": 0.22854892909526825, + "learning_rate": 0.00043938743460304275, + "loss": 1.8415, + "step": 2678 + }, + { + "epoch": 0.26162109375, + "grad_norm": 0.27040791511535645, + "learning_rate": 0.0004393388711889264, + "loss": 1.9017, + "step": 2679 + }, + { + "epoch": 0.26171875, + "grad_norm": 0.2952374219894409, + "learning_rate": 0.0004392902913588128, + "loss": 1.9406, + "step": 2680 + }, + { + "epoch": 0.26181640625, + "grad_norm": 0.27148759365081787, + "learning_rate": 0.0004392416951175544, + "loss": 1.8999, + "step": 2681 + }, + { + "epoch": 0.2619140625, + "grad_norm": 0.24533145129680634, + "learning_rate": 0.00043919308247000575, + "loss": 1.8835, + "step": 2682 + }, + { + "epoch": 0.26201171875, + "grad_norm": 0.2174772322177887, + "learning_rate": 0.0004391444534210229, + "loss": 1.8424, + "step": 2683 + }, + { + "epoch": 0.262109375, + "grad_norm": 0.19030643999576569, + "learning_rate": 0.0004390958079754633, + "loss": 1.8736, + "step": 2684 + }, + { + "epoch": 0.26220703125, + "grad_norm": 0.26530101895332336, + "learning_rate": 0.0004390471461381862, + "loss": 1.8988, + "step": 2685 + }, + { + "epoch": 0.2623046875, + "grad_norm": 0.24065977334976196, + "learning_rate": 0.0004389984679140525, + "loss": 1.9052, + "step": 2686 + }, + { + "epoch": 0.26240234375, + "grad_norm": 0.2812477946281433, + "learning_rate": 0.00043894977330792486, + "loss": 1.8987, + "step": 2687 + }, + { + "epoch": 0.2625, + "grad_norm": 0.2505251169204712, + "learning_rate": 0.00043890106232466724, + "loss": 1.8922, + "step": 2688 + }, + { + "epoch": 0.26259765625, + "grad_norm": 0.2238491326570511, + "learning_rate": 0.0004388523349691455, + "loss": 1.9183, + "step": 2689 + }, + { + "epoch": 0.2626953125, + "grad_norm": 0.26334211230278015, + "learning_rate": 0.00043880359124622714, + "loss": 1.8899, + "step": 2690 + }, + { + "epoch": 0.26279296875, + "grad_norm": 0.24818876385688782, + "learning_rate": 0.0004387548311607812, + "loss": 1.9063, + "step": 2691 + }, + { + "epoch": 0.262890625, + "grad_norm": 0.22725768387317657, + "learning_rate": 0.0004387060547176784, + "loss": 1.8828, + "step": 2692 + }, + { + "epoch": 0.26298828125, + "grad_norm": 0.2095717042684555, + "learning_rate": 0.00043865726192179096, + "loss": 1.8658, + "step": 2693 + }, + { + "epoch": 0.2630859375, + "grad_norm": 0.2227015644311905, + "learning_rate": 0.000438608452777993, + "loss": 1.9546, + "step": 2694 + }, + { + "epoch": 0.26318359375, + "grad_norm": 0.20986926555633545, + "learning_rate": 0.00043855962729116, + "loss": 1.878, + "step": 2695 + }, + { + "epoch": 0.26328125, + "grad_norm": 0.24293240904808044, + "learning_rate": 0.00043851078546616924, + "loss": 1.889, + "step": 2696 + }, + { + "epoch": 0.26337890625, + "grad_norm": 0.23340186476707458, + "learning_rate": 0.0004384619273078996, + "loss": 1.9285, + "step": 2697 + }, + { + "epoch": 0.2634765625, + "grad_norm": 0.22993077337741852, + "learning_rate": 0.00043841305282123157, + "loss": 1.9004, + "step": 2698 + }, + { + "epoch": 0.26357421875, + "grad_norm": 0.25231555104255676, + "learning_rate": 0.00043836416201104727, + "loss": 1.886, + "step": 2699 + }, + { + "epoch": 0.263671875, + "grad_norm": 0.18800829350948334, + "learning_rate": 0.0004383152548822304, + "loss": 1.9054, + "step": 2700 + }, + { + "epoch": 0.26376953125, + "grad_norm": 0.31697878241539, + "learning_rate": 0.0004382663314396665, + "loss": 1.8387, + "step": 2701 + }, + { + "epoch": 0.2638671875, + "grad_norm": 0.31109264492988586, + "learning_rate": 0.00043821739168824246, + "loss": 1.922, + "step": 2702 + }, + { + "epoch": 0.26396484375, + "grad_norm": 0.23077085614204407, + "learning_rate": 0.00043816843563284697, + "loss": 1.8771, + "step": 2703 + }, + { + "epoch": 0.2640625, + "grad_norm": 0.31502625346183777, + "learning_rate": 0.0004381194632783703, + "loss": 1.8655, + "step": 2704 + }, + { + "epoch": 0.26416015625, + "grad_norm": 0.24782267212867737, + "learning_rate": 0.00043807047462970434, + "loss": 1.8778, + "step": 2705 + }, + { + "epoch": 0.2642578125, + "grad_norm": 0.25926321744918823, + "learning_rate": 0.00043802146969174267, + "loss": 1.8437, + "step": 2706 + }, + { + "epoch": 0.26435546875, + "grad_norm": 0.27319055795669556, + "learning_rate": 0.0004379724484693804, + "loss": 1.8433, + "step": 2707 + }, + { + "epoch": 0.264453125, + "grad_norm": 0.23524881899356842, + "learning_rate": 0.0004379234109675143, + "loss": 1.9148, + "step": 2708 + }, + { + "epoch": 0.26455078125, + "grad_norm": 0.28222718834877014, + "learning_rate": 0.0004378743571910429, + "loss": 1.9035, + "step": 2709 + }, + { + "epoch": 0.2646484375, + "grad_norm": 0.271083801984787, + "learning_rate": 0.00043782528714486613, + "loss": 1.8691, + "step": 2710 + }, + { + "epoch": 0.26474609375, + "grad_norm": 0.23110386729240417, + "learning_rate": 0.0004377762008338856, + "loss": 1.8794, + "step": 2711 + }, + { + "epoch": 0.26484375, + "grad_norm": 0.24147644639015198, + "learning_rate": 0.0004377270982630048, + "loss": 1.8577, + "step": 2712 + }, + { + "epoch": 0.26494140625, + "grad_norm": 0.23246973752975464, + "learning_rate": 0.0004376779794371284, + "loss": 1.8876, + "step": 2713 + }, + { + "epoch": 0.2650390625, + "grad_norm": 0.23443640768527985, + "learning_rate": 0.00043762884436116315, + "loss": 1.8693, + "step": 2714 + }, + { + "epoch": 0.26513671875, + "grad_norm": 0.2223389744758606, + "learning_rate": 0.00043757969304001704, + "loss": 1.8308, + "step": 2715 + }, + { + "epoch": 0.265234375, + "grad_norm": 0.25246211886405945, + "learning_rate": 0.0004375305254785999, + "loss": 1.8723, + "step": 2716 + }, + { + "epoch": 0.26533203125, + "grad_norm": 0.2547222077846527, + "learning_rate": 0.0004374813416818232, + "loss": 1.8826, + "step": 2717 + }, + { + "epoch": 0.2654296875, + "grad_norm": 0.2161521464586258, + "learning_rate": 0.0004374321416545999, + "loss": 1.9017, + "step": 2718 + }, + { + "epoch": 0.26552734375, + "grad_norm": 0.2493196278810501, + "learning_rate": 0.0004373829254018447, + "loss": 1.86, + "step": 2719 + }, + { + "epoch": 0.265625, + "grad_norm": 0.2375672310590744, + "learning_rate": 0.00043733369292847386, + "loss": 1.8784, + "step": 2720 + }, + { + "epoch": 0.26572265625, + "grad_norm": 0.34022462368011475, + "learning_rate": 0.00043728444423940516, + "loss": 1.8915, + "step": 2721 + }, + { + "epoch": 0.2658203125, + "grad_norm": 0.25120794773101807, + "learning_rate": 0.0004372351793395582, + "loss": 1.8894, + "step": 2722 + }, + { + "epoch": 0.26591796875, + "grad_norm": 0.2032475620508194, + "learning_rate": 0.0004371858982338542, + "loss": 1.8942, + "step": 2723 + }, + { + "epoch": 0.266015625, + "grad_norm": 0.2685857117176056, + "learning_rate": 0.00043713660092721573, + "loss": 1.9001, + "step": 2724 + }, + { + "epoch": 0.26611328125, + "grad_norm": 0.296835333108902, + "learning_rate": 0.00043708728742456723, + "loss": 1.8994, + "step": 2725 + }, + { + "epoch": 0.2662109375, + "grad_norm": 0.27104151248931885, + "learning_rate": 0.00043703795773083467, + "loss": 1.9119, + "step": 2726 + }, + { + "epoch": 0.26630859375, + "grad_norm": 0.25494450330734253, + "learning_rate": 0.0004369886118509457, + "loss": 1.9188, + "step": 2727 + }, + { + "epoch": 0.26640625, + "grad_norm": 0.3688213527202606, + "learning_rate": 0.0004369392497898294, + "loss": 1.9021, + "step": 2728 + }, + { + "epoch": 0.26650390625, + "grad_norm": 0.24721217155456543, + "learning_rate": 0.00043688987155241676, + "loss": 1.8895, + "step": 2729 + }, + { + "epoch": 0.2666015625, + "grad_norm": 0.290102481842041, + "learning_rate": 0.0004368404771436402, + "loss": 1.8565, + "step": 2730 + }, + { + "epoch": 0.26669921875, + "grad_norm": 0.3468306064605713, + "learning_rate": 0.0004367910665684338, + "loss": 1.8795, + "step": 2731 + }, + { + "epoch": 0.266796875, + "grad_norm": 0.23682735860347748, + "learning_rate": 0.00043674163983173304, + "loss": 1.8749, + "step": 2732 + }, + { + "epoch": 0.26689453125, + "grad_norm": 0.24279217422008514, + "learning_rate": 0.00043669219693847543, + "loss": 1.887, + "step": 2733 + }, + { + "epoch": 0.2669921875, + "grad_norm": 0.25867339968681335, + "learning_rate": 0.0004366427378935998, + "loss": 1.8708, + "step": 2734 + }, + { + "epoch": 0.26708984375, + "grad_norm": 0.28321221470832825, + "learning_rate": 0.0004365932627020467, + "loss": 1.9077, + "step": 2735 + }, + { + "epoch": 0.2671875, + "grad_norm": 0.2243577092885971, + "learning_rate": 0.00043654377136875827, + "loss": 1.8902, + "step": 2736 + }, + { + "epoch": 0.26728515625, + "grad_norm": 0.27794283628463745, + "learning_rate": 0.0004364942638986782, + "loss": 1.8659, + "step": 2737 + }, + { + "epoch": 0.2673828125, + "grad_norm": 0.2588692903518677, + "learning_rate": 0.00043644474029675193, + "loss": 1.8494, + "step": 2738 + }, + { + "epoch": 0.26748046875, + "grad_norm": 0.22354848682880402, + "learning_rate": 0.00043639520056792627, + "loss": 1.895, + "step": 2739 + }, + { + "epoch": 0.267578125, + "grad_norm": 0.3032642900943756, + "learning_rate": 0.00043634564471715, + "loss": 1.8643, + "step": 2740 + }, + { + "epoch": 0.26767578125, + "grad_norm": 0.25967633724212646, + "learning_rate": 0.0004362960727493732, + "loss": 1.8712, + "step": 2741 + }, + { + "epoch": 0.2677734375, + "grad_norm": 0.2637724280357361, + "learning_rate": 0.0004362464846695476, + "loss": 1.8512, + "step": 2742 + }, + { + "epoch": 0.26787109375, + "grad_norm": 0.2793140709400177, + "learning_rate": 0.00043619688048262677, + "loss": 1.9086, + "step": 2743 + }, + { + "epoch": 0.26796875, + "grad_norm": 0.2188546061515808, + "learning_rate": 0.00043614726019356565, + "loss": 1.8649, + "step": 2744 + }, + { + "epoch": 0.26806640625, + "grad_norm": 0.21244989335536957, + "learning_rate": 0.00043609762380732083, + "loss": 1.8744, + "step": 2745 + }, + { + "epoch": 0.2681640625, + "grad_norm": 0.2601993978023529, + "learning_rate": 0.00043604797132885064, + "loss": 1.8649, + "step": 2746 + }, + { + "epoch": 0.26826171875, + "grad_norm": 0.2278032749891281, + "learning_rate": 0.00043599830276311487, + "loss": 1.8662, + "step": 2747 + }, + { + "epoch": 0.268359375, + "grad_norm": 0.24005573987960815, + "learning_rate": 0.000435948618115075, + "loss": 1.8916, + "step": 2748 + }, + { + "epoch": 0.26845703125, + "grad_norm": 0.28929322957992554, + "learning_rate": 0.00043589891738969393, + "loss": 1.9566, + "step": 2749 + }, + { + "epoch": 0.2685546875, + "grad_norm": 0.21914543211460114, + "learning_rate": 0.00043584920059193654, + "loss": 1.9043, + "step": 2750 + }, + { + "epoch": 0.26865234375, + "grad_norm": 0.24009595811367035, + "learning_rate": 0.000435799467726769, + "loss": 1.8493, + "step": 2751 + }, + { + "epoch": 0.26875, + "grad_norm": 0.20415683090686798, + "learning_rate": 0.0004357497187991593, + "loss": 1.8954, + "step": 2752 + }, + { + "epoch": 0.26884765625, + "grad_norm": 0.1941215842962265, + "learning_rate": 0.0004356999538140766, + "loss": 1.8634, + "step": 2753 + }, + { + "epoch": 0.2689453125, + "grad_norm": 0.2202758938074112, + "learning_rate": 0.0004356501727764923, + "loss": 1.8706, + "step": 2754 + }, + { + "epoch": 0.26904296875, + "grad_norm": 0.21078895032405853, + "learning_rate": 0.000435600375691379, + "loss": 1.8942, + "step": 2755 + }, + { + "epoch": 0.269140625, + "grad_norm": 0.21176692843437195, + "learning_rate": 0.00043555056256371083, + "loss": 1.8814, + "step": 2756 + }, + { + "epoch": 0.26923828125, + "grad_norm": 0.20595595240592957, + "learning_rate": 0.00043550073339846394, + "loss": 1.8919, + "step": 2757 + }, + { + "epoch": 0.2693359375, + "grad_norm": 0.2053648680448532, + "learning_rate": 0.00043545088820061564, + "loss": 1.9387, + "step": 2758 + }, + { + "epoch": 0.26943359375, + "grad_norm": 0.2367970496416092, + "learning_rate": 0.00043540102697514513, + "loss": 1.8863, + "step": 2759 + }, + { + "epoch": 0.26953125, + "grad_norm": 0.1842896044254303, + "learning_rate": 0.0004353511497270329, + "loss": 1.8805, + "step": 2760 + }, + { + "epoch": 0.26962890625, + "grad_norm": 0.25769177079200745, + "learning_rate": 0.00043530125646126154, + "loss": 1.8833, + "step": 2761 + }, + { + "epoch": 0.2697265625, + "grad_norm": 0.2760721743106842, + "learning_rate": 0.0004352513471828148, + "loss": 1.8924, + "step": 2762 + }, + { + "epoch": 0.26982421875, + "grad_norm": 0.17177268862724304, + "learning_rate": 0.0004352014218966781, + "loss": 1.8441, + "step": 2763 + }, + { + "epoch": 0.269921875, + "grad_norm": 0.22797243297100067, + "learning_rate": 0.0004351514806078387, + "loss": 1.8753, + "step": 2764 + }, + { + "epoch": 0.27001953125, + "grad_norm": 0.22956639528274536, + "learning_rate": 0.0004351015233212851, + "loss": 1.9303, + "step": 2765 + }, + { + "epoch": 0.2701171875, + "grad_norm": 0.20838238298892975, + "learning_rate": 0.0004350515500420077, + "loss": 1.8689, + "step": 2766 + }, + { + "epoch": 0.27021484375, + "grad_norm": 0.2128564864397049, + "learning_rate": 0.00043500156077499834, + "loss": 1.9195, + "step": 2767 + }, + { + "epoch": 0.2703125, + "grad_norm": 0.21279101073741913, + "learning_rate": 0.00043495155552525074, + "loss": 1.8944, + "step": 2768 + }, + { + "epoch": 0.27041015625, + "grad_norm": 0.22925125062465668, + "learning_rate": 0.00043490153429775964, + "loss": 1.8632, + "step": 2769 + }, + { + "epoch": 0.2705078125, + "grad_norm": 0.23544587194919586, + "learning_rate": 0.00043485149709752193, + "loss": 1.9066, + "step": 2770 + }, + { + "epoch": 0.27060546875, + "grad_norm": 0.22419407963752747, + "learning_rate": 0.00043480144392953567, + "loss": 1.8416, + "step": 2771 + }, + { + "epoch": 0.270703125, + "grad_norm": 0.21770869195461273, + "learning_rate": 0.00043475137479880103, + "loss": 1.9117, + "step": 2772 + }, + { + "epoch": 0.27080078125, + "grad_norm": 0.2358229011297226, + "learning_rate": 0.0004347012897103192, + "loss": 1.9187, + "step": 2773 + }, + { + "epoch": 0.2708984375, + "grad_norm": 0.30673930048942566, + "learning_rate": 0.00043465118866909346, + "loss": 1.8928, + "step": 2774 + }, + { + "epoch": 0.27099609375, + "grad_norm": 0.21168282628059387, + "learning_rate": 0.00043460107168012837, + "loss": 1.9002, + "step": 2775 + }, + { + "epoch": 0.27109375, + "grad_norm": 0.22865557670593262, + "learning_rate": 0.0004345509387484301, + "loss": 1.8956, + "step": 2776 + }, + { + "epoch": 0.27119140625, + "grad_norm": 0.3253214955329895, + "learning_rate": 0.00043450078987900654, + "loss": 1.9049, + "step": 2777 + }, + { + "epoch": 0.2712890625, + "grad_norm": 0.2774486541748047, + "learning_rate": 0.00043445062507686713, + "loss": 1.8817, + "step": 2778 + }, + { + "epoch": 0.27138671875, + "grad_norm": 0.22928157448768616, + "learning_rate": 0.00043440044434702287, + "loss": 1.9167, + "step": 2779 + }, + { + "epoch": 0.271484375, + "grad_norm": 0.2905370891094208, + "learning_rate": 0.0004343502476944864, + "loss": 1.8434, + "step": 2780 + }, + { + "epoch": 0.27158203125, + "grad_norm": 0.19285933673381805, + "learning_rate": 0.0004343000351242719, + "loss": 1.8287, + "step": 2781 + }, + { + "epoch": 0.2716796875, + "grad_norm": 0.25820720195770264, + "learning_rate": 0.0004342498066413951, + "loss": 1.8488, + "step": 2782 + }, + { + "epoch": 0.27177734375, + "grad_norm": 0.2616766095161438, + "learning_rate": 0.00043419956225087354, + "loss": 1.8627, + "step": 2783 + }, + { + "epoch": 0.271875, + "grad_norm": 0.18560141324996948, + "learning_rate": 0.000434149301957726, + "loss": 1.8272, + "step": 2784 + }, + { + "epoch": 0.27197265625, + "grad_norm": 0.3137165307998657, + "learning_rate": 0.0004340990257669732, + "loss": 1.8952, + "step": 2785 + }, + { + "epoch": 0.2720703125, + "grad_norm": 0.231571763753891, + "learning_rate": 0.00043404873368363726, + "loss": 1.9044, + "step": 2786 + }, + { + "epoch": 0.27216796875, + "grad_norm": 0.2827242314815521, + "learning_rate": 0.00043399842571274175, + "loss": 1.8984, + "step": 2787 + }, + { + "epoch": 0.272265625, + "grad_norm": 0.2350980043411255, + "learning_rate": 0.0004339481018593121, + "loss": 1.8902, + "step": 2788 + }, + { + "epoch": 0.27236328125, + "grad_norm": 0.2104049175977707, + "learning_rate": 0.00043389776212837537, + "loss": 1.8853, + "step": 2789 + }, + { + "epoch": 0.2724609375, + "grad_norm": 0.3253656327724457, + "learning_rate": 0.0004338474065249598, + "loss": 1.8982, + "step": 2790 + }, + { + "epoch": 0.27255859375, + "grad_norm": 0.2520630359649658, + "learning_rate": 0.0004337970350540956, + "loss": 1.8806, + "step": 2791 + }, + { + "epoch": 0.27265625, + "grad_norm": 0.2999001145362854, + "learning_rate": 0.0004337466477208145, + "loss": 1.9165, + "step": 2792 + }, + { + "epoch": 0.27275390625, + "grad_norm": 0.20815780758857727, + "learning_rate": 0.0004336962445301495, + "loss": 1.9357, + "step": 2793 + }, + { + "epoch": 0.2728515625, + "grad_norm": 0.24039196968078613, + "learning_rate": 0.0004336458254871357, + "loss": 1.84, + "step": 2794 + }, + { + "epoch": 0.27294921875, + "grad_norm": 0.24862824380397797, + "learning_rate": 0.0004335953905968094, + "loss": 1.8991, + "step": 2795 + }, + { + "epoch": 0.273046875, + "grad_norm": 0.2109254151582718, + "learning_rate": 0.0004335449398642086, + "loss": 1.9077, + "step": 2796 + }, + { + "epoch": 0.27314453125, + "grad_norm": 0.27603134512901306, + "learning_rate": 0.00043349447329437286, + "loss": 1.8703, + "step": 2797 + }, + { + "epoch": 0.2732421875, + "grad_norm": 0.20653647184371948, + "learning_rate": 0.00043344399089234343, + "loss": 1.9106, + "step": 2798 + }, + { + "epoch": 0.27333984375, + "grad_norm": 0.26815375685691833, + "learning_rate": 0.000433393492663163, + "loss": 1.8934, + "step": 2799 + }, + { + "epoch": 0.2734375, + "grad_norm": 0.2597907781600952, + "learning_rate": 0.0004333429786118758, + "loss": 1.8716, + "step": 2800 + }, + { + "epoch": 0.27353515625, + "grad_norm": 0.2918902337551117, + "learning_rate": 0.00043329244874352785, + "loss": 1.8949, + "step": 2801 + }, + { + "epoch": 0.2736328125, + "grad_norm": 0.24110248684883118, + "learning_rate": 0.0004332419030631667, + "loss": 1.91, + "step": 2802 + }, + { + "epoch": 0.27373046875, + "grad_norm": 0.26264944672584534, + "learning_rate": 0.00043319134157584127, + "loss": 1.8656, + "step": 2803 + }, + { + "epoch": 0.273828125, + "grad_norm": 0.30166885256767273, + "learning_rate": 0.00043314076428660237, + "loss": 1.896, + "step": 2804 + }, + { + "epoch": 0.27392578125, + "grad_norm": 0.22887808084487915, + "learning_rate": 0.000433090171200502, + "loss": 1.8776, + "step": 2805 + }, + { + "epoch": 0.2740234375, + "grad_norm": 0.3068729043006897, + "learning_rate": 0.00043303956232259406, + "loss": 1.9091, + "step": 2806 + }, + { + "epoch": 0.27412109375, + "grad_norm": 0.20829269289970398, + "learning_rate": 0.0004329889376579341, + "loss": 1.8863, + "step": 2807 + }, + { + "epoch": 0.27421875, + "grad_norm": 0.3112751245498657, + "learning_rate": 0.00043293829721157876, + "loss": 1.8886, + "step": 2808 + }, + { + "epoch": 0.27431640625, + "grad_norm": 0.2697749137878418, + "learning_rate": 0.00043288764098858677, + "loss": 1.8382, + "step": 2809 + }, + { + "epoch": 0.2744140625, + "grad_norm": 0.27988728880882263, + "learning_rate": 0.0004328369689940182, + "loss": 1.8885, + "step": 2810 + }, + { + "epoch": 0.27451171875, + "grad_norm": 0.2760201394557953, + "learning_rate": 0.0004327862812329348, + "loss": 1.8547, + "step": 2811 + }, + { + "epoch": 0.274609375, + "grad_norm": 0.2351469248533249, + "learning_rate": 0.0004327355777103997, + "loss": 1.9064, + "step": 2812 + }, + { + "epoch": 0.27470703125, + "grad_norm": 0.2867816686630249, + "learning_rate": 0.0004326848584314779, + "loss": 1.9231, + "step": 2813 + }, + { + "epoch": 0.2748046875, + "grad_norm": 0.2675870954990387, + "learning_rate": 0.00043263412340123556, + "loss": 1.9473, + "step": 2814 + }, + { + "epoch": 0.27490234375, + "grad_norm": 0.2948281466960907, + "learning_rate": 0.0004325833726247409, + "loss": 1.8853, + "step": 2815 + }, + { + "epoch": 0.275, + "grad_norm": 0.2507423460483551, + "learning_rate": 0.0004325326061070634, + "loss": 1.8954, + "step": 2816 + }, + { + "epoch": 0.27509765625, + "grad_norm": 0.328163743019104, + "learning_rate": 0.0004324818238532741, + "loss": 1.8834, + "step": 2817 + }, + { + "epoch": 0.2751953125, + "grad_norm": 0.24358054995536804, + "learning_rate": 0.0004324310258684458, + "loss": 1.9065, + "step": 2818 + }, + { + "epoch": 0.27529296875, + "grad_norm": 0.2406340092420578, + "learning_rate": 0.0004323802121576527, + "loss": 1.8833, + "step": 2819 + }, + { + "epoch": 0.275390625, + "grad_norm": 0.22684375941753387, + "learning_rate": 0.00043232938272597067, + "loss": 1.8548, + "step": 2820 + }, + { + "epoch": 0.27548828125, + "grad_norm": 0.26202863454818726, + "learning_rate": 0.00043227853757847715, + "loss": 1.8859, + "step": 2821 + }, + { + "epoch": 0.2755859375, + "grad_norm": 0.3058609962463379, + "learning_rate": 0.00043222767672025105, + "loss": 1.8672, + "step": 2822 + }, + { + "epoch": 0.27568359375, + "grad_norm": 0.20576193928718567, + "learning_rate": 0.000432176800156373, + "loss": 1.8819, + "step": 2823 + }, + { + "epoch": 0.27578125, + "grad_norm": 0.2687084376811981, + "learning_rate": 0.000432125907891925, + "loss": 1.8617, + "step": 2824 + }, + { + "epoch": 0.27587890625, + "grad_norm": 0.29834067821502686, + "learning_rate": 0.00043207499993199084, + "loss": 1.9432, + "step": 2825 + }, + { + "epoch": 0.2759765625, + "grad_norm": 0.2804504930973053, + "learning_rate": 0.0004320240762816558, + "loss": 1.9183, + "step": 2826 + }, + { + "epoch": 0.27607421875, + "grad_norm": 0.2958560883998871, + "learning_rate": 0.00043197313694600666, + "loss": 1.8613, + "step": 2827 + }, + { + "epoch": 0.276171875, + "grad_norm": 0.19655700027942657, + "learning_rate": 0.0004319221819301317, + "loss": 1.9211, + "step": 2828 + }, + { + "epoch": 0.27626953125, + "grad_norm": 0.24349285662174225, + "learning_rate": 0.00043187121123912104, + "loss": 1.8844, + "step": 2829 + }, + { + "epoch": 0.2763671875, + "grad_norm": 0.21005815267562866, + "learning_rate": 0.0004318202248780661, + "loss": 1.8959, + "step": 2830 + }, + { + "epoch": 0.27646484375, + "grad_norm": 0.20665428042411804, + "learning_rate": 0.00043176922285205997, + "loss": 1.9001, + "step": 2831 + }, + { + "epoch": 0.2765625, + "grad_norm": 0.2562062740325928, + "learning_rate": 0.00043171820516619744, + "loss": 1.8792, + "step": 2832 + }, + { + "epoch": 0.27666015625, + "grad_norm": 0.18265512585639954, + "learning_rate": 0.00043166717182557455, + "loss": 1.9092, + "step": 2833 + }, + { + "epoch": 0.2767578125, + "grad_norm": 0.2603815793991089, + "learning_rate": 0.0004316161228352891, + "loss": 1.9011, + "step": 2834 + }, + { + "epoch": 0.27685546875, + "grad_norm": 0.25518473982810974, + "learning_rate": 0.0004315650582004405, + "loss": 1.8912, + "step": 2835 + }, + { + "epoch": 0.276953125, + "grad_norm": 0.20616333186626434, + "learning_rate": 0.00043151397792612963, + "loss": 1.8852, + "step": 2836 + }, + { + "epoch": 0.27705078125, + "grad_norm": 0.21389137208461761, + "learning_rate": 0.00043146288201745893, + "loss": 1.8471, + "step": 2837 + }, + { + "epoch": 0.2771484375, + "grad_norm": 0.20006729662418365, + "learning_rate": 0.00043141177047953253, + "loss": 1.8815, + "step": 2838 + }, + { + "epoch": 0.27724609375, + "grad_norm": 0.22058548033237457, + "learning_rate": 0.0004313606433174559, + "loss": 1.879, + "step": 2839 + }, + { + "epoch": 0.27734375, + "grad_norm": 0.21773791313171387, + "learning_rate": 0.0004313095005363362, + "loss": 1.8815, + "step": 2840 + }, + { + "epoch": 0.27744140625, + "grad_norm": 0.20097985863685608, + "learning_rate": 0.0004312583421412822, + "loss": 1.8748, + "step": 2841 + }, + { + "epoch": 0.2775390625, + "grad_norm": 0.22170375287532806, + "learning_rate": 0.00043120716813740414, + "loss": 1.8611, + "step": 2842 + }, + { + "epoch": 0.27763671875, + "grad_norm": 0.20943517982959747, + "learning_rate": 0.00043115597852981383, + "loss": 1.8566, + "step": 2843 + }, + { + "epoch": 0.277734375, + "grad_norm": 0.22400376200675964, + "learning_rate": 0.0004311047733236247, + "loss": 1.8881, + "step": 2844 + }, + { + "epoch": 0.27783203125, + "grad_norm": 0.2491447776556015, + "learning_rate": 0.0004310535525239517, + "loss": 1.902, + "step": 2845 + }, + { + "epoch": 0.2779296875, + "grad_norm": 0.2509153187274933, + "learning_rate": 0.0004310023161359113, + "loss": 1.8923, + "step": 2846 + }, + { + "epoch": 0.27802734375, + "grad_norm": 0.25616857409477234, + "learning_rate": 0.00043095106416462153, + "loss": 1.9018, + "step": 2847 + }, + { + "epoch": 0.278125, + "grad_norm": 0.2445756494998932, + "learning_rate": 0.0004308997966152022, + "loss": 1.8785, + "step": 2848 + }, + { + "epoch": 0.27822265625, + "grad_norm": 0.2224990576505661, + "learning_rate": 0.00043084851349277413, + "loss": 1.8688, + "step": 2849 + }, + { + "epoch": 0.2783203125, + "grad_norm": 0.2594734728336334, + "learning_rate": 0.0004307972148024604, + "loss": 1.8407, + "step": 2850 + }, + { + "epoch": 0.27841796875, + "grad_norm": 0.2550109326839447, + "learning_rate": 0.00043074590054938513, + "loss": 1.9068, + "step": 2851 + }, + { + "epoch": 0.278515625, + "grad_norm": 0.24641066789627075, + "learning_rate": 0.0004306945707386741, + "loss": 1.8777, + "step": 2852 + }, + { + "epoch": 0.27861328125, + "grad_norm": 0.230759397149086, + "learning_rate": 0.0004306432253754549, + "loss": 1.8931, + "step": 2853 + }, + { + "epoch": 0.2787109375, + "grad_norm": 0.25458577275276184, + "learning_rate": 0.0004305918644648562, + "loss": 1.9012, + "step": 2854 + }, + { + "epoch": 0.27880859375, + "grad_norm": 0.23450514674186707, + "learning_rate": 0.00043054048801200877, + "loss": 1.8963, + "step": 2855 + }, + { + "epoch": 0.27890625, + "grad_norm": 0.2019408941268921, + "learning_rate": 0.0004304890960220446, + "loss": 1.9174, + "step": 2856 + }, + { + "epoch": 0.27900390625, + "grad_norm": 0.24433976411819458, + "learning_rate": 0.0004304376885000972, + "loss": 1.8657, + "step": 2857 + }, + { + "epoch": 0.2791015625, + "grad_norm": 0.23637981712818146, + "learning_rate": 0.0004303862654513017, + "loss": 1.8746, + "step": 2858 + }, + { + "epoch": 0.27919921875, + "grad_norm": 0.23872153460979462, + "learning_rate": 0.000430334826880795, + "loss": 1.8968, + "step": 2859 + }, + { + "epoch": 0.279296875, + "grad_norm": 0.30130165815353394, + "learning_rate": 0.0004302833727937151, + "loss": 1.8725, + "step": 2860 + }, + { + "epoch": 0.27939453125, + "grad_norm": 0.26016566157341003, + "learning_rate": 0.00043023190319520205, + "loss": 1.9012, + "step": 2861 + }, + { + "epoch": 0.2794921875, + "grad_norm": 0.22780142724514008, + "learning_rate": 0.0004301804180903971, + "loss": 1.829, + "step": 2862 + }, + { + "epoch": 0.27958984375, + "grad_norm": 0.23210953176021576, + "learning_rate": 0.00043012891748444313, + "loss": 1.893, + "step": 2863 + }, + { + "epoch": 0.2796875, + "grad_norm": 0.2725372612476349, + "learning_rate": 0.00043007740138248466, + "loss": 1.8397, + "step": 2864 + }, + { + "epoch": 0.27978515625, + "grad_norm": 0.23973391950130463, + "learning_rate": 0.0004300258697896675, + "loss": 1.8947, + "step": 2865 + }, + { + "epoch": 0.2798828125, + "grad_norm": 0.2913033664226532, + "learning_rate": 0.0004299743227111395, + "loss": 1.863, + "step": 2866 + }, + { + "epoch": 0.27998046875, + "grad_norm": 0.23055680096149445, + "learning_rate": 0.0004299227601520496, + "loss": 1.8307, + "step": 2867 + }, + { + "epoch": 0.280078125, + "grad_norm": 0.26329392194747925, + "learning_rate": 0.00042987118211754844, + "loss": 1.8753, + "step": 2868 + }, + { + "epoch": 0.28017578125, + "grad_norm": 0.3165866732597351, + "learning_rate": 0.0004298195886127882, + "loss": 1.8636, + "step": 2869 + }, + { + "epoch": 0.2802734375, + "grad_norm": 0.2055087685585022, + "learning_rate": 0.0004297679796429226, + "loss": 1.8981, + "step": 2870 + }, + { + "epoch": 0.28037109375, + "grad_norm": 0.23618242144584656, + "learning_rate": 0.00042971635521310705, + "loss": 1.935, + "step": 2871 + }, + { + "epoch": 0.28046875, + "grad_norm": 0.21628020703792572, + "learning_rate": 0.0004296647153284982, + "loss": 1.8618, + "step": 2872 + }, + { + "epoch": 0.28056640625, + "grad_norm": 0.24062804877758026, + "learning_rate": 0.0004296130599942546, + "loss": 1.8927, + "step": 2873 + }, + { + "epoch": 0.2806640625, + "grad_norm": 0.25762608647346497, + "learning_rate": 0.00042956138921553596, + "loss": 1.8769, + "step": 2874 + }, + { + "epoch": 0.28076171875, + "grad_norm": 0.2140209972858429, + "learning_rate": 0.00042950970299750386, + "loss": 1.8871, + "step": 2875 + }, + { + "epoch": 0.280859375, + "grad_norm": 0.24165524542331696, + "learning_rate": 0.0004294580013453213, + "loss": 1.9134, + "step": 2876 + }, + { + "epoch": 0.28095703125, + "grad_norm": 0.21578289568424225, + "learning_rate": 0.00042940628426415265, + "loss": 1.868, + "step": 2877 + }, + { + "epoch": 0.2810546875, + "grad_norm": 0.24762646853923798, + "learning_rate": 0.00042935455175916414, + "loss": 1.8629, + "step": 2878 + }, + { + "epoch": 0.28115234375, + "grad_norm": 0.22082191705703735, + "learning_rate": 0.0004293028038355235, + "loss": 1.8574, + "step": 2879 + }, + { + "epoch": 0.28125, + "grad_norm": 0.22337709367275238, + "learning_rate": 0.0004292510404983997, + "loss": 1.903, + "step": 2880 + }, + { + "epoch": 0.28134765625, + "grad_norm": 0.23375888168811798, + "learning_rate": 0.0004291992617529635, + "loss": 1.8598, + "step": 2881 + }, + { + "epoch": 0.2814453125, + "grad_norm": 0.23466967046260834, + "learning_rate": 0.00042914746760438707, + "loss": 1.9082, + "step": 2882 + }, + { + "epoch": 0.28154296875, + "grad_norm": 0.2091575413942337, + "learning_rate": 0.0004290956580578443, + "loss": 1.8389, + "step": 2883 + }, + { + "epoch": 0.281640625, + "grad_norm": 0.27531692385673523, + "learning_rate": 0.0004290438331185105, + "loss": 1.876, + "step": 2884 + }, + { + "epoch": 0.28173828125, + "grad_norm": 0.24806031584739685, + "learning_rate": 0.0004289919927915624, + "loss": 1.8586, + "step": 2885 + }, + { + "epoch": 0.2818359375, + "grad_norm": 0.19965526461601257, + "learning_rate": 0.00042894013708217845, + "loss": 1.8679, + "step": 2886 + }, + { + "epoch": 0.28193359375, + "grad_norm": 0.2199181616306305, + "learning_rate": 0.00042888826599553863, + "loss": 1.9465, + "step": 2887 + }, + { + "epoch": 0.28203125, + "grad_norm": 0.2529730200767517, + "learning_rate": 0.0004288363795368244, + "loss": 1.9187, + "step": 2888 + }, + { + "epoch": 0.28212890625, + "grad_norm": 0.299998939037323, + "learning_rate": 0.00042878447771121865, + "loss": 1.8727, + "step": 2889 + }, + { + "epoch": 0.2822265625, + "grad_norm": 0.30152031779289246, + "learning_rate": 0.000428732560523906, + "loss": 1.9078, + "step": 2890 + }, + { + "epoch": 0.28232421875, + "grad_norm": 0.28815585374832153, + "learning_rate": 0.0004286806279800726, + "loss": 1.9571, + "step": 2891 + }, + { + "epoch": 0.282421875, + "grad_norm": 0.2583948075771332, + "learning_rate": 0.00042862868008490583, + "loss": 1.9134, + "step": 2892 + }, + { + "epoch": 0.28251953125, + "grad_norm": 0.23865336179733276, + "learning_rate": 0.000428576716843595, + "loss": 1.8894, + "step": 2893 + }, + { + "epoch": 0.2826171875, + "grad_norm": 0.2828342020511627, + "learning_rate": 0.00042852473826133076, + "loss": 1.8806, + "step": 2894 + }, + { + "epoch": 0.28271484375, + "grad_norm": 0.23666423559188843, + "learning_rate": 0.0004284727443433052, + "loss": 1.8645, + "step": 2895 + }, + { + "epoch": 0.2828125, + "grad_norm": 0.23653465509414673, + "learning_rate": 0.0004284207350947122, + "loss": 1.8928, + "step": 2896 + }, + { + "epoch": 0.28291015625, + "grad_norm": 0.21783040463924408, + "learning_rate": 0.0004283687105207469, + "loss": 1.9072, + "step": 2897 + }, + { + "epoch": 0.2830078125, + "grad_norm": 0.2457730621099472, + "learning_rate": 0.0004283166706266061, + "loss": 1.8927, + "step": 2898 + }, + { + "epoch": 0.28310546875, + "grad_norm": 0.19717064499855042, + "learning_rate": 0.0004282646154174882, + "loss": 1.8715, + "step": 2899 + }, + { + "epoch": 0.283203125, + "grad_norm": 0.22203285992145538, + "learning_rate": 0.00042821254489859307, + "loss": 1.8398, + "step": 2900 + }, + { + "epoch": 0.28330078125, + "grad_norm": 0.21506066620349884, + "learning_rate": 0.00042816045907512207, + "loss": 1.9277, + "step": 2901 + }, + { + "epoch": 0.2833984375, + "grad_norm": 0.268706351518631, + "learning_rate": 0.00042810835795227797, + "loss": 1.8684, + "step": 2902 + }, + { + "epoch": 0.28349609375, + "grad_norm": 0.19960474967956543, + "learning_rate": 0.0004280562415352654, + "loss": 1.9536, + "step": 2903 + }, + { + "epoch": 0.28359375, + "grad_norm": 0.216094970703125, + "learning_rate": 0.00042800410982929025, + "loss": 1.8531, + "step": 2904 + }, + { + "epoch": 0.28369140625, + "grad_norm": 0.25142574310302734, + "learning_rate": 0.00042795196283956007, + "loss": 1.8876, + "step": 2905 + }, + { + "epoch": 0.2837890625, + "grad_norm": 0.24804849922657013, + "learning_rate": 0.0004278998005712837, + "loss": 1.8942, + "step": 2906 + }, + { + "epoch": 0.28388671875, + "grad_norm": 0.2586076557636261, + "learning_rate": 0.000427847623029672, + "loss": 1.9142, + "step": 2907 + }, + { + "epoch": 0.283984375, + "grad_norm": 0.2308054268360138, + "learning_rate": 0.0004277954302199368, + "loss": 1.8652, + "step": 2908 + }, + { + "epoch": 0.28408203125, + "grad_norm": 0.24985018372535706, + "learning_rate": 0.00042774322214729173, + "loss": 1.911, + "step": 2909 + }, + { + "epoch": 0.2841796875, + "grad_norm": 0.3349003493785858, + "learning_rate": 0.00042769099881695194, + "loss": 1.955, + "step": 2910 + }, + { + "epoch": 0.28427734375, + "grad_norm": 0.2801406979560852, + "learning_rate": 0.0004276387602341341, + "loss": 1.878, + "step": 2911 + }, + { + "epoch": 0.284375, + "grad_norm": 0.2786557674407959, + "learning_rate": 0.0004275865064040564, + "loss": 1.9241, + "step": 2912 + }, + { + "epoch": 0.28447265625, + "grad_norm": 0.3159611225128174, + "learning_rate": 0.0004275342373319385, + "loss": 1.8619, + "step": 2913 + }, + { + "epoch": 0.2845703125, + "grad_norm": 0.3373079001903534, + "learning_rate": 0.0004274819530230016, + "loss": 1.8697, + "step": 2914 + }, + { + "epoch": 0.28466796875, + "grad_norm": 0.22254876792430878, + "learning_rate": 0.00042742965348246855, + "loss": 1.9184, + "step": 2915 + }, + { + "epoch": 0.284765625, + "grad_norm": 0.2410246878862381, + "learning_rate": 0.00042737733871556335, + "loss": 1.9146, + "step": 2916 + }, + { + "epoch": 0.28486328125, + "grad_norm": 0.2892058491706848, + "learning_rate": 0.0004273250087275121, + "loss": 1.8803, + "step": 2917 + }, + { + "epoch": 0.2849609375, + "grad_norm": 0.27044740319252014, + "learning_rate": 0.0004272726635235419, + "loss": 1.8896, + "step": 2918 + }, + { + "epoch": 0.28505859375, + "grad_norm": 0.18537931144237518, + "learning_rate": 0.0004272203031088816, + "loss": 1.8655, + "step": 2919 + }, + { + "epoch": 0.28515625, + "grad_norm": 0.307284414768219, + "learning_rate": 0.00042716792748876157, + "loss": 1.9111, + "step": 2920 + }, + { + "epoch": 0.28525390625, + "grad_norm": 0.30498889088630676, + "learning_rate": 0.00042711553666841375, + "loss": 1.8797, + "step": 2921 + }, + { + "epoch": 0.2853515625, + "grad_norm": 0.22686442732810974, + "learning_rate": 0.0004270631306530714, + "loss": 1.8822, + "step": 2922 + }, + { + "epoch": 0.28544921875, + "grad_norm": 0.27979257702827454, + "learning_rate": 0.0004270107094479694, + "loss": 1.9223, + "step": 2923 + }, + { + "epoch": 0.285546875, + "grad_norm": 0.2563009262084961, + "learning_rate": 0.00042695827305834434, + "loss": 1.8761, + "step": 2924 + }, + { + "epoch": 0.28564453125, + "grad_norm": 0.2728877067565918, + "learning_rate": 0.000426905821489434, + "loss": 1.9284, + "step": 2925 + }, + { + "epoch": 0.2857421875, + "grad_norm": 0.23300801217556, + "learning_rate": 0.00042685335474647773, + "loss": 1.8814, + "step": 2926 + }, + { + "epoch": 0.28583984375, + "grad_norm": 0.3054649233818054, + "learning_rate": 0.0004268008728347168, + "loss": 1.8466, + "step": 2927 + }, + { + "epoch": 0.2859375, + "grad_norm": 0.210355743765831, + "learning_rate": 0.00042674837575939334, + "loss": 1.8857, + "step": 2928 + }, + { + "epoch": 0.28603515625, + "grad_norm": 0.25826016068458557, + "learning_rate": 0.00042669586352575155, + "loss": 1.8571, + "step": 2929 + }, + { + "epoch": 0.2861328125, + "grad_norm": 0.231784388422966, + "learning_rate": 0.00042664333613903695, + "loss": 1.8342, + "step": 2930 + }, + { + "epoch": 0.28623046875, + "grad_norm": 0.22335542738437653, + "learning_rate": 0.00042659079360449654, + "loss": 1.8573, + "step": 2931 + }, + { + "epoch": 0.286328125, + "grad_norm": 0.22986647486686707, + "learning_rate": 0.0004265382359273788, + "loss": 1.9117, + "step": 2932 + }, + { + "epoch": 0.28642578125, + "grad_norm": 0.27009549736976624, + "learning_rate": 0.0004264856631129338, + "loss": 1.8988, + "step": 2933 + }, + { + "epoch": 0.2865234375, + "grad_norm": 0.23258306086063385, + "learning_rate": 0.000426433075166413, + "loss": 1.8654, + "step": 2934 + }, + { + "epoch": 0.28662109375, + "grad_norm": 0.23832274973392487, + "learning_rate": 0.0004263804720930697, + "loss": 1.8759, + "step": 2935 + }, + { + "epoch": 0.28671875, + "grad_norm": 0.31919005513191223, + "learning_rate": 0.00042632785389815824, + "loss": 1.8901, + "step": 2936 + }, + { + "epoch": 0.28681640625, + "grad_norm": 0.2530565559864044, + "learning_rate": 0.00042627522058693483, + "loss": 1.8673, + "step": 2937 + }, + { + "epoch": 0.2869140625, + "grad_norm": 0.26637616753578186, + "learning_rate": 0.00042622257216465715, + "loss": 1.8791, + "step": 2938 + }, + { + "epoch": 0.28701171875, + "grad_norm": 0.2281026691198349, + "learning_rate": 0.00042616990863658416, + "loss": 1.8664, + "step": 2939 + }, + { + "epoch": 0.287109375, + "grad_norm": 0.23909759521484375, + "learning_rate": 0.0004261172300079766, + "loss": 1.8193, + "step": 2940 + }, + { + "epoch": 0.28720703125, + "grad_norm": 0.22489655017852783, + "learning_rate": 0.0004260645362840965, + "loss": 1.8929, + "step": 2941 + }, + { + "epoch": 0.2873046875, + "grad_norm": 0.20088250935077667, + "learning_rate": 0.00042601182747020744, + "loss": 1.8831, + "step": 2942 + }, + { + "epoch": 0.28740234375, + "grad_norm": 0.3041893541812897, + "learning_rate": 0.0004259591035715748, + "loss": 1.8802, + "step": 2943 + }, + { + "epoch": 0.2875, + "grad_norm": 0.19707295298576355, + "learning_rate": 0.00042590636459346505, + "loss": 1.8729, + "step": 2944 + }, + { + "epoch": 0.28759765625, + "grad_norm": 0.26492106914520264, + "learning_rate": 0.00042585361054114636, + "loss": 1.9214, + "step": 2945 + }, + { + "epoch": 0.2876953125, + "grad_norm": 0.23438039422035217, + "learning_rate": 0.00042580084141988846, + "loss": 1.9079, + "step": 2946 + }, + { + "epoch": 0.28779296875, + "grad_norm": 0.2153705209493637, + "learning_rate": 0.0004257480572349625, + "loss": 1.8923, + "step": 2947 + }, + { + "epoch": 0.287890625, + "grad_norm": 0.2112143486738205, + "learning_rate": 0.000425695257991641, + "loss": 1.844, + "step": 2948 + }, + { + "epoch": 0.28798828125, + "grad_norm": 0.25230222940444946, + "learning_rate": 0.00042564244369519837, + "loss": 1.8469, + "step": 2949 + }, + { + "epoch": 0.2880859375, + "grad_norm": 0.2979770600795746, + "learning_rate": 0.00042558961435091013, + "loss": 1.8678, + "step": 2950 + }, + { + "epoch": 0.28818359375, + "grad_norm": 0.21866418421268463, + "learning_rate": 0.00042553676996405363, + "loss": 1.8903, + "step": 2951 + }, + { + "epoch": 0.28828125, + "grad_norm": 0.23139160871505737, + "learning_rate": 0.0004254839105399073, + "loss": 1.8967, + "step": 2952 + }, + { + "epoch": 0.28837890625, + "grad_norm": 0.22103898227214813, + "learning_rate": 0.0004254310360837516, + "loss": 1.8925, + "step": 2953 + }, + { + "epoch": 0.2884765625, + "grad_norm": 0.21849390864372253, + "learning_rate": 0.00042537814660086797, + "loss": 1.8718, + "step": 2954 + }, + { + "epoch": 0.28857421875, + "grad_norm": 0.23933055996894836, + "learning_rate": 0.0004253252420965398, + "loss": 1.8892, + "step": 2955 + }, + { + "epoch": 0.288671875, + "grad_norm": 0.20929864048957825, + "learning_rate": 0.00042527232257605173, + "loss": 1.8613, + "step": 2956 + }, + { + "epoch": 0.28876953125, + "grad_norm": 0.24336673319339752, + "learning_rate": 0.0004252193880446899, + "loss": 1.9355, + "step": 2957 + }, + { + "epoch": 0.2888671875, + "grad_norm": 0.2018166482448578, + "learning_rate": 0.000425166438507742, + "loss": 1.8684, + "step": 2958 + }, + { + "epoch": 0.28896484375, + "grad_norm": 0.1958925426006317, + "learning_rate": 0.0004251134739704973, + "loss": 1.8739, + "step": 2959 + }, + { + "epoch": 0.2890625, + "grad_norm": 0.2154635787010193, + "learning_rate": 0.0004250604944382464, + "loss": 1.8547, + "step": 2960 + }, + { + "epoch": 0.28916015625, + "grad_norm": 0.21988588571548462, + "learning_rate": 0.0004250074999162816, + "loss": 1.8857, + "step": 2961 + }, + { + "epoch": 0.2892578125, + "grad_norm": 0.1953074336051941, + "learning_rate": 0.0004249544904098964, + "loss": 1.8928, + "step": 2962 + }, + { + "epoch": 0.28935546875, + "grad_norm": 0.18395958840847015, + "learning_rate": 0.00042490146592438614, + "loss": 1.8739, + "step": 2963 + }, + { + "epoch": 0.289453125, + "grad_norm": 0.21299080550670624, + "learning_rate": 0.0004248484264650474, + "loss": 1.8542, + "step": 2964 + }, + { + "epoch": 0.28955078125, + "grad_norm": 0.23363280296325684, + "learning_rate": 0.0004247953720371784, + "loss": 1.8985, + "step": 2965 + }, + { + "epoch": 0.2896484375, + "grad_norm": 0.23393043875694275, + "learning_rate": 0.00042474230264607886, + "loss": 1.899, + "step": 2966 + }, + { + "epoch": 0.28974609375, + "grad_norm": 0.202012300491333, + "learning_rate": 0.0004246892182970499, + "loss": 1.8999, + "step": 2967 + }, + { + "epoch": 0.28984375, + "grad_norm": 0.23654010891914368, + "learning_rate": 0.000424636118995394, + "loss": 1.8621, + "step": 2968 + }, + { + "epoch": 0.28994140625, + "grad_norm": 0.18624065816402435, + "learning_rate": 0.0004245830047464156, + "loss": 1.8794, + "step": 2969 + }, + { + "epoch": 0.2900390625, + "grad_norm": 0.22178705036640167, + "learning_rate": 0.00042452987555542025, + "loss": 1.9132, + "step": 2970 + }, + { + "epoch": 0.29013671875, + "grad_norm": 0.2666608691215515, + "learning_rate": 0.0004244767314277149, + "loss": 1.842, + "step": 2971 + }, + { + "epoch": 0.290234375, + "grad_norm": 0.21921314299106598, + "learning_rate": 0.0004244235723686084, + "loss": 1.8707, + "step": 2972 + }, + { + "epoch": 0.29033203125, + "grad_norm": 0.26558271050453186, + "learning_rate": 0.00042437039838341087, + "loss": 1.8919, + "step": 2973 + }, + { + "epoch": 0.2904296875, + "grad_norm": 0.2301444709300995, + "learning_rate": 0.00042431720947743373, + "loss": 1.8578, + "step": 2974 + }, + { + "epoch": 0.29052734375, + "grad_norm": 0.21512138843536377, + "learning_rate": 0.0004242640056559902, + "loss": 1.8769, + "step": 2975 + }, + { + "epoch": 0.290625, + "grad_norm": 0.20394618809223175, + "learning_rate": 0.00042421078692439496, + "loss": 1.9135, + "step": 2976 + }, + { + "epoch": 0.29072265625, + "grad_norm": 0.22402624785900116, + "learning_rate": 0.000424157553287964, + "loss": 1.9198, + "step": 2977 + }, + { + "epoch": 0.2908203125, + "grad_norm": 0.2281763255596161, + "learning_rate": 0.0004241043047520147, + "loss": 1.8726, + "step": 2978 + }, + { + "epoch": 0.29091796875, + "grad_norm": 0.22465281188488007, + "learning_rate": 0.0004240510413218665, + "loss": 1.8684, + "step": 2979 + }, + { + "epoch": 0.291015625, + "grad_norm": 0.26606932282447815, + "learning_rate": 0.00042399776300283963, + "loss": 1.9309, + "step": 2980 + }, + { + "epoch": 0.29111328125, + "grad_norm": 0.2602376639842987, + "learning_rate": 0.0004239444698002563, + "loss": 1.819, + "step": 2981 + }, + { + "epoch": 0.2912109375, + "grad_norm": 0.2876492440700531, + "learning_rate": 0.00042389116171943995, + "loss": 1.8752, + "step": 2982 + }, + { + "epoch": 0.29130859375, + "grad_norm": 0.24838502705097198, + "learning_rate": 0.00042383783876571556, + "loss": 1.8708, + "step": 2983 + }, + { + "epoch": 0.29140625, + "grad_norm": 0.2866203188896179, + "learning_rate": 0.0004237845009444097, + "loss": 1.8873, + "step": 2984 + }, + { + "epoch": 0.29150390625, + "grad_norm": 0.3191507160663605, + "learning_rate": 0.0004237311482608503, + "loss": 1.8515, + "step": 2985 + }, + { + "epoch": 0.2916015625, + "grad_norm": 0.306711345911026, + "learning_rate": 0.00042367778072036684, + "loss": 1.8751, + "step": 2986 + }, + { + "epoch": 0.29169921875, + "grad_norm": 0.23548844456672668, + "learning_rate": 0.00042362439832829027, + "loss": 1.867, + "step": 2987 + }, + { + "epoch": 0.291796875, + "grad_norm": 0.27832359075546265, + "learning_rate": 0.00042357100108995297, + "loss": 1.8451, + "step": 2988 + }, + { + "epoch": 0.29189453125, + "grad_norm": 0.35388869047164917, + "learning_rate": 0.0004235175890106889, + "loss": 1.8578, + "step": 2989 + }, + { + "epoch": 0.2919921875, + "grad_norm": 0.21929793059825897, + "learning_rate": 0.00042346416209583336, + "loss": 1.8308, + "step": 2990 + }, + { + "epoch": 0.29208984375, + "grad_norm": 0.2917014956474304, + "learning_rate": 0.00042341072035072345, + "loss": 1.888, + "step": 2991 + }, + { + "epoch": 0.2921875, + "grad_norm": 0.2988588511943817, + "learning_rate": 0.0004233572637806972, + "loss": 1.8833, + "step": 2992 + }, + { + "epoch": 0.29228515625, + "grad_norm": 0.24378128349781036, + "learning_rate": 0.0004233037923910947, + "loss": 1.8409, + "step": 2993 + }, + { + "epoch": 0.2923828125, + "grad_norm": 0.2123136818408966, + "learning_rate": 0.00042325030618725726, + "loss": 1.8501, + "step": 2994 + }, + { + "epoch": 0.29248046875, + "grad_norm": 0.2910783886909485, + "learning_rate": 0.0004231968051745275, + "loss": 1.8987, + "step": 2995 + }, + { + "epoch": 0.292578125, + "grad_norm": 0.3038376271724701, + "learning_rate": 0.00042314328935824994, + "loss": 1.9294, + "step": 2996 + }, + { + "epoch": 0.29267578125, + "grad_norm": 0.2226715087890625, + "learning_rate": 0.0004230897587437701, + "loss": 1.8779, + "step": 2997 + }, + { + "epoch": 0.2927734375, + "grad_norm": 0.24885967373847961, + "learning_rate": 0.0004230362133364354, + "loss": 1.854, + "step": 2998 + }, + { + "epoch": 0.29287109375, + "grad_norm": 0.22683629393577576, + "learning_rate": 0.0004229826531415943, + "loss": 1.8815, + "step": 2999 + }, + { + "epoch": 0.29296875, + "grad_norm": 0.2502784729003906, + "learning_rate": 0.00042292907816459736, + "loss": 1.8787, + "step": 3000 + }, + { + "epoch": 0.29306640625, + "grad_norm": 0.22668708860874176, + "learning_rate": 0.00042287548841079593, + "loss": 1.867, + "step": 3001 + }, + { + "epoch": 0.2931640625, + "grad_norm": 0.2151390165090561, + "learning_rate": 0.0004228218838855433, + "loss": 1.9048, + "step": 3002 + }, + { + "epoch": 0.29326171875, + "grad_norm": 0.2484009712934494, + "learning_rate": 0.0004227682645941941, + "loss": 1.8894, + "step": 3003 + }, + { + "epoch": 0.293359375, + "grad_norm": 0.2573185861110687, + "learning_rate": 0.0004227146305421043, + "loss": 1.8898, + "step": 3004 + }, + { + "epoch": 0.29345703125, + "grad_norm": 0.23028187453746796, + "learning_rate": 0.0004226609817346316, + "loss": 1.8766, + "step": 3005 + }, + { + "epoch": 0.2935546875, + "grad_norm": 0.2626796364784241, + "learning_rate": 0.00042260731817713495, + "loss": 1.9076, + "step": 3006 + }, + { + "epoch": 0.29365234375, + "grad_norm": 0.23662570118904114, + "learning_rate": 0.0004225536398749749, + "loss": 1.8151, + "step": 3007 + }, + { + "epoch": 0.29375, + "grad_norm": 0.20050571858882904, + "learning_rate": 0.00042249994683351346, + "loss": 1.86, + "step": 3008 + }, + { + "epoch": 0.29384765625, + "grad_norm": 0.2129492163658142, + "learning_rate": 0.00042244623905811395, + "loss": 1.9022, + "step": 3009 + }, + { + "epoch": 0.2939453125, + "grad_norm": 0.19077609479427338, + "learning_rate": 0.00042239251655414153, + "loss": 1.8774, + "step": 3010 + }, + { + "epoch": 0.29404296875, + "grad_norm": 0.21290133893489838, + "learning_rate": 0.00042233877932696247, + "loss": 1.8751, + "step": 3011 + }, + { + "epoch": 0.294140625, + "grad_norm": 0.23995377123355865, + "learning_rate": 0.0004222850273819447, + "loss": 1.9138, + "step": 3012 + }, + { + "epoch": 0.29423828125, + "grad_norm": 0.2578481435775757, + "learning_rate": 0.00042223126072445745, + "loss": 1.8922, + "step": 3013 + }, + { + "epoch": 0.2943359375, + "grad_norm": 0.2387392222881317, + "learning_rate": 0.00042217747935987166, + "loss": 1.8643, + "step": 3014 + }, + { + "epoch": 0.29443359375, + "grad_norm": 0.2922835946083069, + "learning_rate": 0.00042212368329355953, + "loss": 1.81, + "step": 3015 + }, + { + "epoch": 0.29453125, + "grad_norm": 0.20787885785102844, + "learning_rate": 0.00042206987253089477, + "loss": 1.8931, + "step": 3016 + }, + { + "epoch": 0.29462890625, + "grad_norm": 0.2633446753025055, + "learning_rate": 0.0004220160470772528, + "loss": 1.8675, + "step": 3017 + }, + { + "epoch": 0.2947265625, + "grad_norm": 0.1937829703092575, + "learning_rate": 0.0004219622069380101, + "loss": 1.8966, + "step": 3018 + }, + { + "epoch": 0.29482421875, + "grad_norm": 0.26633307337760925, + "learning_rate": 0.0004219083521185449, + "loss": 1.8361, + "step": 3019 + }, + { + "epoch": 0.294921875, + "grad_norm": 0.2277202308177948, + "learning_rate": 0.00042185448262423673, + "loss": 1.8753, + "step": 3020 + }, + { + "epoch": 0.29501953125, + "grad_norm": 0.19210180640220642, + "learning_rate": 0.00042180059846046685, + "loss": 1.8922, + "step": 3021 + }, + { + "epoch": 0.2951171875, + "grad_norm": 0.20748457312583923, + "learning_rate": 0.0004217466996326178, + "loss": 1.898, + "step": 3022 + }, + { + "epoch": 0.29521484375, + "grad_norm": 0.1865435540676117, + "learning_rate": 0.0004216927861460733, + "loss": 1.9276, + "step": 3023 + }, + { + "epoch": 0.2953125, + "grad_norm": 0.2428085207939148, + "learning_rate": 0.0004216388580062192, + "loss": 1.8543, + "step": 3024 + }, + { + "epoch": 0.29541015625, + "grad_norm": 0.22671908140182495, + "learning_rate": 0.00042158491521844234, + "loss": 1.8916, + "step": 3025 + }, + { + "epoch": 0.2955078125, + "grad_norm": 0.22087739408016205, + "learning_rate": 0.00042153095778813096, + "loss": 1.8361, + "step": 3026 + }, + { + "epoch": 0.29560546875, + "grad_norm": 0.21374981105327606, + "learning_rate": 0.00042147698572067506, + "loss": 1.8736, + "step": 3027 + }, + { + "epoch": 0.295703125, + "grad_norm": 0.2175641655921936, + "learning_rate": 0.000421422999021466, + "loss": 1.8394, + "step": 3028 + }, + { + "epoch": 0.29580078125, + "grad_norm": 0.28677627444267273, + "learning_rate": 0.0004213689976958965, + "loss": 1.8548, + "step": 3029 + }, + { + "epoch": 0.2958984375, + "grad_norm": 0.2653857469558716, + "learning_rate": 0.00042131498174936084, + "loss": 1.873, + "step": 3030 + }, + { + "epoch": 0.29599609375, + "grad_norm": 0.23550917208194733, + "learning_rate": 0.00042126095118725474, + "loss": 1.8541, + "step": 3031 + }, + { + "epoch": 0.29609375, + "grad_norm": 0.2541399598121643, + "learning_rate": 0.00042120690601497537, + "loss": 1.8607, + "step": 3032 + }, + { + "epoch": 0.29619140625, + "grad_norm": 0.2753686010837555, + "learning_rate": 0.00042115284623792137, + "loss": 1.9019, + "step": 3033 + }, + { + "epoch": 0.2962890625, + "grad_norm": 0.2758835554122925, + "learning_rate": 0.0004210987718614928, + "loss": 1.9017, + "step": 3034 + }, + { + "epoch": 0.29638671875, + "grad_norm": 0.22177983820438385, + "learning_rate": 0.0004210446828910913, + "loss": 1.8971, + "step": 3035 + }, + { + "epoch": 0.296484375, + "grad_norm": 0.22884881496429443, + "learning_rate": 0.00042099057933211976, + "loss": 1.8347, + "step": 3036 + }, + { + "epoch": 0.29658203125, + "grad_norm": 0.27076879143714905, + "learning_rate": 0.00042093646118998275, + "loss": 1.86, + "step": 3037 + }, + { + "epoch": 0.2966796875, + "grad_norm": 0.32488006353378296, + "learning_rate": 0.0004208823284700861, + "loss": 1.9228, + "step": 3038 + }, + { + "epoch": 0.29677734375, + "grad_norm": 0.22296027839183807, + "learning_rate": 0.00042082818117783726, + "loss": 1.9108, + "step": 3039 + }, + { + "epoch": 0.296875, + "grad_norm": 0.29426977038383484, + "learning_rate": 0.00042077401931864505, + "loss": 1.8609, + "step": 3040 + }, + { + "epoch": 0.29697265625, + "grad_norm": 0.3135233521461487, + "learning_rate": 0.00042071984289791976, + "loss": 1.8902, + "step": 3041 + }, + { + "epoch": 0.2970703125, + "grad_norm": 0.3182580769062042, + "learning_rate": 0.00042066565192107314, + "loss": 1.8717, + "step": 3042 + }, + { + "epoch": 0.29716796875, + "grad_norm": 0.2791113257408142, + "learning_rate": 0.00042061144639351833, + "loss": 1.9108, + "step": 3043 + }, + { + "epoch": 0.297265625, + "grad_norm": 0.24948930740356445, + "learning_rate": 0.00042055722632067006, + "loss": 1.8509, + "step": 3044 + }, + { + "epoch": 0.29736328125, + "grad_norm": 0.270920991897583, + "learning_rate": 0.0004205029917079444, + "loss": 1.8868, + "step": 3045 + }, + { + "epoch": 0.2974609375, + "grad_norm": 0.2594306766986847, + "learning_rate": 0.0004204487425607589, + "loss": 1.8583, + "step": 3046 + }, + { + "epoch": 0.29755859375, + "grad_norm": 0.24974964559078217, + "learning_rate": 0.00042039447888453255, + "loss": 1.8907, + "step": 3047 + }, + { + "epoch": 0.29765625, + "grad_norm": 0.25526463985443115, + "learning_rate": 0.00042034020068468583, + "loss": 1.904, + "step": 3048 + }, + { + "epoch": 0.29775390625, + "grad_norm": 0.24909386038780212, + "learning_rate": 0.00042028590796664073, + "loss": 1.8768, + "step": 3049 + }, + { + "epoch": 0.2978515625, + "grad_norm": 0.20612211525440216, + "learning_rate": 0.0004202316007358205, + "loss": 1.9036, + "step": 3050 + }, + { + "epoch": 0.29794921875, + "grad_norm": 0.26715099811553955, + "learning_rate": 0.00042017727899764995, + "loss": 1.9315, + "step": 3051 + }, + { + "epoch": 0.298046875, + "grad_norm": 0.21301884949207306, + "learning_rate": 0.00042012294275755544, + "loss": 1.8762, + "step": 3052 + }, + { + "epoch": 0.29814453125, + "grad_norm": 0.23274759948253632, + "learning_rate": 0.00042006859202096464, + "loss": 1.9061, + "step": 3053 + }, + { + "epoch": 0.2982421875, + "grad_norm": 0.23647235333919525, + "learning_rate": 0.00042001422679330665, + "loss": 1.8344, + "step": 3054 + }, + { + "epoch": 0.29833984375, + "grad_norm": 0.22753413021564484, + "learning_rate": 0.00041995984708001217, + "loss": 1.8933, + "step": 3055 + }, + { + "epoch": 0.2984375, + "grad_norm": 0.25154316425323486, + "learning_rate": 0.0004199054528865131, + "loss": 1.8551, + "step": 3056 + }, + { + "epoch": 0.29853515625, + "grad_norm": 0.22566568851470947, + "learning_rate": 0.0004198510442182431, + "loss": 1.8548, + "step": 3057 + }, + { + "epoch": 0.2986328125, + "grad_norm": 0.21213006973266602, + "learning_rate": 0.000419796621080637, + "loss": 1.8801, + "step": 3058 + }, + { + "epoch": 0.29873046875, + "grad_norm": 0.2567862570285797, + "learning_rate": 0.00041974218347913127, + "loss": 1.8701, + "step": 3059 + }, + { + "epoch": 0.298828125, + "grad_norm": 0.25658440589904785, + "learning_rate": 0.0004196877314191637, + "loss": 1.9191, + "step": 3060 + }, + { + "epoch": 0.29892578125, + "grad_norm": 0.21651284396648407, + "learning_rate": 0.0004196332649061736, + "loss": 1.8633, + "step": 3061 + }, + { + "epoch": 0.2990234375, + "grad_norm": 0.23801001906394958, + "learning_rate": 0.00041957878394560167, + "loss": 1.9149, + "step": 3062 + }, + { + "epoch": 0.29912109375, + "grad_norm": 0.21144436299800873, + "learning_rate": 0.00041952428854289005, + "loss": 1.8981, + "step": 3063 + }, + { + "epoch": 0.29921875, + "grad_norm": 0.2747276723384857, + "learning_rate": 0.0004194697787034824, + "loss": 1.8633, + "step": 3064 + }, + { + "epoch": 0.29931640625, + "grad_norm": 0.2292105257511139, + "learning_rate": 0.0004194152544328237, + "loss": 1.8304, + "step": 3065 + }, + { + "epoch": 0.2994140625, + "grad_norm": 0.19762180745601654, + "learning_rate": 0.0004193607157363606, + "loss": 1.8328, + "step": 3066 + }, + { + "epoch": 0.29951171875, + "grad_norm": 0.27750471234321594, + "learning_rate": 0.0004193061626195408, + "loss": 1.8736, + "step": 3067 + }, + { + "epoch": 0.299609375, + "grad_norm": 0.19942927360534668, + "learning_rate": 0.0004192515950878138, + "loss": 1.8863, + "step": 3068 + }, + { + "epoch": 0.29970703125, + "grad_norm": 0.23817166686058044, + "learning_rate": 0.0004191970131466304, + "loss": 1.8638, + "step": 3069 + }, + { + "epoch": 0.2998046875, + "grad_norm": 0.20811405777931213, + "learning_rate": 0.0004191424168014429, + "loss": 1.9001, + "step": 3070 + }, + { + "epoch": 0.29990234375, + "grad_norm": 0.23758384585380554, + "learning_rate": 0.00041908780605770497, + "loss": 1.9237, + "step": 3071 + }, + { + "epoch": 0.3, + "grad_norm": 0.20853745937347412, + "learning_rate": 0.0004190331809208717, + "loss": 1.9003, + "step": 3072 + }, + { + "epoch": 0.30009765625, + "grad_norm": 0.23026727139949799, + "learning_rate": 0.00041897854139639963, + "loss": 1.8439, + "step": 3073 + }, + { + "epoch": 0.3001953125, + "grad_norm": 0.1935022622346878, + "learning_rate": 0.00041892388748974687, + "loss": 1.8977, + "step": 3074 + }, + { + "epoch": 0.30029296875, + "grad_norm": 0.22622525691986084, + "learning_rate": 0.0004188692192063727, + "loss": 1.8909, + "step": 3075 + }, + { + "epoch": 0.300390625, + "grad_norm": 0.2203913778066635, + "learning_rate": 0.00041881453655173825, + "loss": 1.8693, + "step": 3076 + }, + { + "epoch": 0.30048828125, + "grad_norm": 0.18365997076034546, + "learning_rate": 0.00041875983953130563, + "loss": 1.8997, + "step": 3077 + }, + { + "epoch": 0.3005859375, + "grad_norm": 0.1894032061100006, + "learning_rate": 0.0004187051281505386, + "loss": 1.884, + "step": 3078 + }, + { + "epoch": 0.30068359375, + "grad_norm": 0.17142687737941742, + "learning_rate": 0.0004186504024149023, + "loss": 1.8654, + "step": 3079 + }, + { + "epoch": 0.30078125, + "grad_norm": 0.20617616176605225, + "learning_rate": 0.00041859566232986355, + "loss": 1.8877, + "step": 3080 + }, + { + "epoch": 0.30087890625, + "grad_norm": 0.2163764387369156, + "learning_rate": 0.0004185409079008903, + "loss": 1.8872, + "step": 3081 + }, + { + "epoch": 0.3009765625, + "grad_norm": 0.22180713713169098, + "learning_rate": 0.0004184861391334519, + "loss": 1.8953, + "step": 3082 + }, + { + "epoch": 0.30107421875, + "grad_norm": 0.24861370027065277, + "learning_rate": 0.00041843135603301945, + "loss": 1.9071, + "step": 3083 + }, + { + "epoch": 0.301171875, + "grad_norm": 0.20080408453941345, + "learning_rate": 0.0004183765586050652, + "loss": 1.8864, + "step": 3084 + }, + { + "epoch": 0.30126953125, + "grad_norm": 0.22699226438999176, + "learning_rate": 0.00041832174685506296, + "loss": 1.8346, + "step": 3085 + }, + { + "epoch": 0.3013671875, + "grad_norm": 0.23934027552604675, + "learning_rate": 0.00041826692078848787, + "loss": 1.9335, + "step": 3086 + }, + { + "epoch": 0.30146484375, + "grad_norm": 0.2329765111207962, + "learning_rate": 0.0004182120804108167, + "loss": 1.8811, + "step": 3087 + }, + { + "epoch": 0.3015625, + "grad_norm": 0.21421007812023163, + "learning_rate": 0.0004181572257275274, + "loss": 1.9029, + "step": 3088 + }, + { + "epoch": 0.30166015625, + "grad_norm": 0.22248250246047974, + "learning_rate": 0.00041810235674409954, + "loss": 1.8565, + "step": 3089 + }, + { + "epoch": 0.3017578125, + "grad_norm": 0.2047787755727768, + "learning_rate": 0.0004180474734660139, + "loss": 1.8361, + "step": 3090 + }, + { + "epoch": 0.30185546875, + "grad_norm": 0.2110084891319275, + "learning_rate": 0.00041799257589875306, + "loss": 1.9514, + "step": 3091 + }, + { + "epoch": 0.301953125, + "grad_norm": 0.2262723594903946, + "learning_rate": 0.0004179376640478007, + "loss": 1.8518, + "step": 3092 + }, + { + "epoch": 0.30205078125, + "grad_norm": 0.23749500513076782, + "learning_rate": 0.0004178827379186419, + "loss": 1.895, + "step": 3093 + }, + { + "epoch": 0.3021484375, + "grad_norm": 0.21547161042690277, + "learning_rate": 0.0004178277975167635, + "loss": 1.8941, + "step": 3094 + }, + { + "epoch": 0.30224609375, + "grad_norm": 0.22931936383247375, + "learning_rate": 0.00041777284284765344, + "loss": 1.8962, + "step": 3095 + }, + { + "epoch": 0.30234375, + "grad_norm": 0.21383483707904816, + "learning_rate": 0.0004177178739168013, + "loss": 1.8861, + "step": 3096 + }, + { + "epoch": 0.30244140625, + "grad_norm": 0.21916463971138, + "learning_rate": 0.00041766289072969786, + "loss": 1.8892, + "step": 3097 + }, + { + "epoch": 0.3025390625, + "grad_norm": 0.23424991965293884, + "learning_rate": 0.00041760789329183554, + "loss": 1.8532, + "step": 3098 + }, + { + "epoch": 0.30263671875, + "grad_norm": 0.2427515983581543, + "learning_rate": 0.0004175528816087081, + "loss": 1.9044, + "step": 3099 + }, + { + "epoch": 0.302734375, + "grad_norm": 0.17762315273284912, + "learning_rate": 0.00041749785568581064, + "loss": 1.8533, + "step": 3100 + }, + { + "epoch": 0.30283203125, + "grad_norm": 0.24818706512451172, + "learning_rate": 0.0004174428155286397, + "loss": 1.8871, + "step": 3101 + }, + { + "epoch": 0.3029296875, + "grad_norm": 0.26211288571357727, + "learning_rate": 0.0004173877611426936, + "loss": 1.8751, + "step": 3102 + }, + { + "epoch": 0.30302734375, + "grad_norm": 0.27161601185798645, + "learning_rate": 0.00041733269253347156, + "loss": 1.9253, + "step": 3103 + }, + { + "epoch": 0.303125, + "grad_norm": 0.2077367603778839, + "learning_rate": 0.00041727760970647446, + "loss": 1.8505, + "step": 3104 + }, + { + "epoch": 0.30322265625, + "grad_norm": 0.19965888559818268, + "learning_rate": 0.00041722251266720467, + "loss": 1.8617, + "step": 3105 + }, + { + "epoch": 0.3033203125, + "grad_norm": 0.2016630917787552, + "learning_rate": 0.0004171674014211658, + "loss": 1.892, + "step": 3106 + }, + { + "epoch": 0.30341796875, + "grad_norm": 0.2344559282064438, + "learning_rate": 0.000417112275973863, + "loss": 1.8858, + "step": 3107 + }, + { + "epoch": 0.303515625, + "grad_norm": 0.21691851317882538, + "learning_rate": 0.00041705713633080285, + "loss": 1.9133, + "step": 3108 + }, + { + "epoch": 0.30361328125, + "grad_norm": 0.1909625083208084, + "learning_rate": 0.00041700198249749326, + "loss": 1.868, + "step": 3109 + }, + { + "epoch": 0.3037109375, + "grad_norm": 0.1911030411720276, + "learning_rate": 0.0004169468144794437, + "loss": 1.8663, + "step": 3110 + }, + { + "epoch": 0.30380859375, + "grad_norm": 0.23135457932949066, + "learning_rate": 0.0004168916322821649, + "loss": 1.8741, + "step": 3111 + }, + { + "epoch": 0.30390625, + "grad_norm": 0.17381919920444489, + "learning_rate": 0.0004168364359111691, + "loss": 1.8367, + "step": 3112 + }, + { + "epoch": 0.30400390625, + "grad_norm": 0.22277741134166718, + "learning_rate": 0.0004167812253719698, + "loss": 1.9282, + "step": 3113 + }, + { + "epoch": 0.3041015625, + "grad_norm": 0.22977705299854279, + "learning_rate": 0.0004167260006700823, + "loss": 1.829, + "step": 3114 + }, + { + "epoch": 0.30419921875, + "grad_norm": 0.17196373641490936, + "learning_rate": 0.0004166707618110228, + "loss": 1.9048, + "step": 3115 + }, + { + "epoch": 0.304296875, + "grad_norm": 0.2016908824443817, + "learning_rate": 0.0004166155088003094, + "loss": 1.89, + "step": 3116 + }, + { + "epoch": 0.30439453125, + "grad_norm": 0.20606733858585358, + "learning_rate": 0.00041656024164346116, + "loss": 1.8202, + "step": 3117 + }, + { + "epoch": 0.3044921875, + "grad_norm": 0.22512635588645935, + "learning_rate": 0.00041650496034599895, + "loss": 1.8737, + "step": 3118 + }, + { + "epoch": 0.30458984375, + "grad_norm": 0.23294253647327423, + "learning_rate": 0.00041644966491344483, + "loss": 1.8862, + "step": 3119 + }, + { + "epoch": 0.3046875, + "grad_norm": 0.19793957471847534, + "learning_rate": 0.0004163943553513222, + "loss": 1.8846, + "step": 3120 + }, + { + "epoch": 0.30478515625, + "grad_norm": 0.22563882172107697, + "learning_rate": 0.00041633903166515623, + "loss": 1.8554, + "step": 3121 + }, + { + "epoch": 0.3048828125, + "grad_norm": 0.20656296610832214, + "learning_rate": 0.00041628369386047313, + "loss": 1.8328, + "step": 3122 + }, + { + "epoch": 0.30498046875, + "grad_norm": 0.2028147280216217, + "learning_rate": 0.00041622834194280067, + "loss": 1.8821, + "step": 3123 + }, + { + "epoch": 0.305078125, + "grad_norm": 0.22198504209518433, + "learning_rate": 0.00041617297591766794, + "loss": 1.8595, + "step": 3124 + }, + { + "epoch": 0.30517578125, + "grad_norm": 0.18492154777050018, + "learning_rate": 0.0004161175957906058, + "loss": 1.8535, + "step": 3125 + }, + { + "epoch": 0.3052734375, + "grad_norm": 0.28677722811698914, + "learning_rate": 0.00041606220156714587, + "loss": 1.8726, + "step": 3126 + }, + { + "epoch": 0.30537109375, + "grad_norm": 0.28656336665153503, + "learning_rate": 0.00041600679325282173, + "loss": 1.8607, + "step": 3127 + }, + { + "epoch": 0.30546875, + "grad_norm": 0.2700391709804535, + "learning_rate": 0.0004159513708531681, + "loss": 1.8796, + "step": 3128 + }, + { + "epoch": 0.30556640625, + "grad_norm": 0.254393607378006, + "learning_rate": 0.00041589593437372135, + "loss": 1.9238, + "step": 3129 + }, + { + "epoch": 0.3056640625, + "grad_norm": 0.21035775542259216, + "learning_rate": 0.000415840483820019, + "loss": 1.9173, + "step": 3130 + }, + { + "epoch": 0.30576171875, + "grad_norm": 0.2469177544116974, + "learning_rate": 0.0004157850191975999, + "loss": 1.8812, + "step": 3131 + }, + { + "epoch": 0.305859375, + "grad_norm": 0.3074721693992615, + "learning_rate": 0.00041572954051200466, + "loss": 1.8538, + "step": 3132 + }, + { + "epoch": 0.30595703125, + "grad_norm": 0.25160425901412964, + "learning_rate": 0.00041567404776877515, + "loss": 1.9504, + "step": 3133 + }, + { + "epoch": 0.3060546875, + "grad_norm": 0.24114885926246643, + "learning_rate": 0.0004156185409734545, + "loss": 1.8791, + "step": 3134 + }, + { + "epoch": 0.30615234375, + "grad_norm": 0.26935282349586487, + "learning_rate": 0.00041556302013158735, + "loss": 1.9038, + "step": 3135 + }, + { + "epoch": 0.30625, + "grad_norm": 0.23060311377048492, + "learning_rate": 0.0004155074852487198, + "loss": 1.8994, + "step": 3136 + }, + { + "epoch": 0.30634765625, + "grad_norm": 0.2560269832611084, + "learning_rate": 0.0004154519363303993, + "loss": 1.858, + "step": 3137 + }, + { + "epoch": 0.3064453125, + "grad_norm": 0.25312769412994385, + "learning_rate": 0.0004153963733821745, + "loss": 1.9092, + "step": 3138 + }, + { + "epoch": 0.30654296875, + "grad_norm": 0.26236045360565186, + "learning_rate": 0.0004153407964095959, + "loss": 1.9356, + "step": 3139 + }, + { + "epoch": 0.306640625, + "grad_norm": 0.24468998610973358, + "learning_rate": 0.0004152852054182151, + "loss": 1.8854, + "step": 3140 + }, + { + "epoch": 0.30673828125, + "grad_norm": 0.21663248538970947, + "learning_rate": 0.000415229600413585, + "loss": 1.7767, + "step": 3141 + }, + { + "epoch": 0.3068359375, + "grad_norm": 0.2483324110507965, + "learning_rate": 0.0004151739814012602, + "loss": 1.876, + "step": 3142 + }, + { + "epoch": 0.30693359375, + "grad_norm": 0.26244667172431946, + "learning_rate": 0.00041511834838679643, + "loss": 1.8613, + "step": 3143 + }, + { + "epoch": 0.30703125, + "grad_norm": 0.23940621316432953, + "learning_rate": 0.00041506270137575105, + "loss": 1.8662, + "step": 3144 + }, + { + "epoch": 0.30712890625, + "grad_norm": 0.2661633789539337, + "learning_rate": 0.00041500704037368254, + "loss": 1.8443, + "step": 3145 + }, + { + "epoch": 0.3072265625, + "grad_norm": 0.24965199828147888, + "learning_rate": 0.0004149513653861512, + "loss": 1.8688, + "step": 3146 + }, + { + "epoch": 0.30732421875, + "grad_norm": 0.25112828612327576, + "learning_rate": 0.00041489567641871827, + "loss": 1.8736, + "step": 3147 + }, + { + "epoch": 0.307421875, + "grad_norm": 0.2768343985080719, + "learning_rate": 0.00041483997347694653, + "loss": 1.8818, + "step": 3148 + }, + { + "epoch": 0.30751953125, + "grad_norm": 0.2519071698188782, + "learning_rate": 0.0004147842565664004, + "loss": 1.8826, + "step": 3149 + }, + { + "epoch": 0.3076171875, + "grad_norm": 0.24958127737045288, + "learning_rate": 0.00041472852569264545, + "loss": 1.8985, + "step": 3150 + }, + { + "epoch": 0.30771484375, + "grad_norm": 0.18655551970005035, + "learning_rate": 0.0004146727808612486, + "loss": 1.8777, + "step": 3151 + }, + { + "epoch": 0.3078125, + "grad_norm": 0.21460086107254028, + "learning_rate": 0.0004146170220777783, + "loss": 1.8659, + "step": 3152 + }, + { + "epoch": 0.30791015625, + "grad_norm": 0.20738814771175385, + "learning_rate": 0.0004145612493478044, + "loss": 1.882, + "step": 3153 + }, + { + "epoch": 0.3080078125, + "grad_norm": 0.2415093630552292, + "learning_rate": 0.00041450546267689817, + "loss": 1.9349, + "step": 3154 + }, + { + "epoch": 0.30810546875, + "grad_norm": 0.1886264681816101, + "learning_rate": 0.0004144496620706321, + "loss": 1.8221, + "step": 3155 + }, + { + "epoch": 0.308203125, + "grad_norm": 0.21637070178985596, + "learning_rate": 0.00041439384753458016, + "loss": 1.886, + "step": 3156 + }, + { + "epoch": 0.30830078125, + "grad_norm": 0.19392646849155426, + "learning_rate": 0.0004143380190743178, + "loss": 1.8497, + "step": 3157 + }, + { + "epoch": 0.3083984375, + "grad_norm": 0.20178459584712982, + "learning_rate": 0.0004142821766954218, + "loss": 1.9139, + "step": 3158 + }, + { + "epoch": 0.30849609375, + "grad_norm": 0.22789818048477173, + "learning_rate": 0.0004142263204034702, + "loss": 1.8608, + "step": 3159 + }, + { + "epoch": 0.30859375, + "grad_norm": 0.19092601537704468, + "learning_rate": 0.00041417045020404267, + "loss": 1.9351, + "step": 3160 + }, + { + "epoch": 0.30869140625, + "grad_norm": 0.2776362895965576, + "learning_rate": 0.0004141145661027201, + "loss": 1.9225, + "step": 3161 + }, + { + "epoch": 0.3087890625, + "grad_norm": 0.2744249999523163, + "learning_rate": 0.0004140586681050848, + "loss": 1.8647, + "step": 3162 + }, + { + "epoch": 0.30888671875, + "grad_norm": 0.2721487283706665, + "learning_rate": 0.0004140027562167205, + "loss": 1.859, + "step": 3163 + }, + { + "epoch": 0.308984375, + "grad_norm": 0.29660922288894653, + "learning_rate": 0.00041394683044321235, + "loss": 1.9029, + "step": 3164 + }, + { + "epoch": 0.30908203125, + "grad_norm": 0.28247225284576416, + "learning_rate": 0.0004138908907901468, + "loss": 1.8388, + "step": 3165 + }, + { + "epoch": 0.3091796875, + "grad_norm": 0.18411090970039368, + "learning_rate": 0.0004138349372631116, + "loss": 1.9018, + "step": 3166 + }, + { + "epoch": 0.30927734375, + "grad_norm": 0.31455278396606445, + "learning_rate": 0.00041377896986769625, + "loss": 1.8912, + "step": 3167 + }, + { + "epoch": 0.309375, + "grad_norm": 0.24204468727111816, + "learning_rate": 0.0004137229886094913, + "loss": 1.9221, + "step": 3168 + }, + { + "epoch": 0.30947265625, + "grad_norm": 0.24073071777820587, + "learning_rate": 0.0004136669934940886, + "loss": 1.8923, + "step": 3169 + }, + { + "epoch": 0.3095703125, + "grad_norm": 0.25627797842025757, + "learning_rate": 0.00041361098452708195, + "loss": 1.8898, + "step": 3170 + }, + { + "epoch": 0.30966796875, + "grad_norm": 0.21789778769016266, + "learning_rate": 0.0004135549617140658, + "loss": 1.8953, + "step": 3171 + }, + { + "epoch": 0.309765625, + "grad_norm": 0.2414090931415558, + "learning_rate": 0.00041349892506063654, + "loss": 1.8784, + "step": 3172 + }, + { + "epoch": 0.30986328125, + "grad_norm": 0.25079676508903503, + "learning_rate": 0.0004134428745723916, + "loss": 1.9243, + "step": 3173 + }, + { + "epoch": 0.3099609375, + "grad_norm": 0.261700838804245, + "learning_rate": 0.00041338681025492995, + "loss": 1.8871, + "step": 3174 + }, + { + "epoch": 0.31005859375, + "grad_norm": 0.2135803997516632, + "learning_rate": 0.000413330732113852, + "loss": 1.8587, + "step": 3175 + }, + { + "epoch": 0.31015625, + "grad_norm": 0.22568845748901367, + "learning_rate": 0.00041327464015475943, + "loss": 1.877, + "step": 3176 + }, + { + "epoch": 0.31025390625, + "grad_norm": 0.2725401520729065, + "learning_rate": 0.0004132185343832553, + "loss": 1.8597, + "step": 3177 + }, + { + "epoch": 0.3103515625, + "grad_norm": 0.24924452602863312, + "learning_rate": 0.00041316241480494404, + "loss": 1.9168, + "step": 3178 + }, + { + "epoch": 0.31044921875, + "grad_norm": 0.24290919303894043, + "learning_rate": 0.00041310628142543157, + "loss": 1.9064, + "step": 3179 + }, + { + "epoch": 0.310546875, + "grad_norm": 0.2438277006149292, + "learning_rate": 0.00041305013425032514, + "loss": 1.8389, + "step": 3180 + }, + { + "epoch": 0.31064453125, + "grad_norm": 0.24335592985153198, + "learning_rate": 0.00041299397328523323, + "loss": 1.8927, + "step": 3181 + }, + { + "epoch": 0.3107421875, + "grad_norm": 0.2722269296646118, + "learning_rate": 0.00041293779853576604, + "loss": 1.8622, + "step": 3182 + }, + { + "epoch": 0.31083984375, + "grad_norm": 0.20185427367687225, + "learning_rate": 0.0004128816100075347, + "loss": 1.8776, + "step": 3183 + }, + { + "epoch": 0.3109375, + "grad_norm": 0.24118156731128693, + "learning_rate": 0.00041282540770615206, + "loss": 1.8891, + "step": 3184 + }, + { + "epoch": 0.31103515625, + "grad_norm": 0.2829108238220215, + "learning_rate": 0.0004127691916372322, + "loss": 1.893, + "step": 3185 + }, + { + "epoch": 0.3111328125, + "grad_norm": 0.20308874547481537, + "learning_rate": 0.00041271296180639064, + "loss": 1.8062, + "step": 3186 + }, + { + "epoch": 0.31123046875, + "grad_norm": 0.26868608593940735, + "learning_rate": 0.00041265671821924424, + "loss": 1.8488, + "step": 3187 + }, + { + "epoch": 0.311328125, + "grad_norm": 0.20011785626411438, + "learning_rate": 0.00041260046088141116, + "loss": 1.901, + "step": 3188 + }, + { + "epoch": 0.31142578125, + "grad_norm": 0.22014351189136505, + "learning_rate": 0.0004125441897985111, + "loss": 1.9075, + "step": 3189 + }, + { + "epoch": 0.3115234375, + "grad_norm": 0.21280397474765778, + "learning_rate": 0.0004124879049761651, + "loss": 1.901, + "step": 3190 + }, + { + "epoch": 0.31162109375, + "grad_norm": 0.19353657960891724, + "learning_rate": 0.00041243160641999527, + "loss": 1.8421, + "step": 3191 + }, + { + "epoch": 0.31171875, + "grad_norm": 0.2556135952472687, + "learning_rate": 0.00041237529413562556, + "loss": 1.8846, + "step": 3192 + }, + { + "epoch": 0.31181640625, + "grad_norm": 0.1932743787765503, + "learning_rate": 0.00041231896812868096, + "loss": 1.8886, + "step": 3193 + }, + { + "epoch": 0.3119140625, + "grad_norm": 0.21884669363498688, + "learning_rate": 0.00041226262840478805, + "loss": 1.923, + "step": 3194 + }, + { + "epoch": 0.31201171875, + "grad_norm": 0.2124539613723755, + "learning_rate": 0.0004122062749695746, + "loss": 1.8899, + "step": 3195 + }, + { + "epoch": 0.312109375, + "grad_norm": 0.20314374566078186, + "learning_rate": 0.0004121499078286697, + "loss": 1.8547, + "step": 3196 + }, + { + "epoch": 0.31220703125, + "grad_norm": 0.24776391685009003, + "learning_rate": 0.0004120935269877043, + "loss": 1.8759, + "step": 3197 + }, + { + "epoch": 0.3123046875, + "grad_norm": 0.1929536908864975, + "learning_rate": 0.0004120371324523098, + "loss": 1.8746, + "step": 3198 + }, + { + "epoch": 0.31240234375, + "grad_norm": 0.21577607095241547, + "learning_rate": 0.00041198072422812, + "loss": 1.8879, + "step": 3199 + }, + { + "epoch": 0.3125, + "grad_norm": 0.19796085357666016, + "learning_rate": 0.00041192430232076926, + "loss": 1.8855, + "step": 3200 + }, + { + "epoch": 0.31259765625, + "grad_norm": 0.24222511053085327, + "learning_rate": 0.00041186786673589384, + "loss": 1.9066, + "step": 3201 + }, + { + "epoch": 0.3126953125, + "grad_norm": 0.24613934755325317, + "learning_rate": 0.00041181141747913106, + "loss": 1.8442, + "step": 3202 + }, + { + "epoch": 0.31279296875, + "grad_norm": 0.2534504532814026, + "learning_rate": 0.0004117549545561197, + "loss": 1.8432, + "step": 3203 + }, + { + "epoch": 0.312890625, + "grad_norm": 0.23667122423648834, + "learning_rate": 0.0004116984779724999, + "loss": 1.8551, + "step": 3204 + }, + { + "epoch": 0.31298828125, + "grad_norm": 0.21642574667930603, + "learning_rate": 0.0004116419877339133, + "loss": 1.9147, + "step": 3205 + }, + { + "epoch": 0.3130859375, + "grad_norm": 0.2664399743080139, + "learning_rate": 0.0004115854838460026, + "loss": 1.9004, + "step": 3206 + }, + { + "epoch": 0.31318359375, + "grad_norm": 0.27308887243270874, + "learning_rate": 0.00041152896631441204, + "loss": 1.8451, + "step": 3207 + }, + { + "epoch": 0.31328125, + "grad_norm": 0.287526398897171, + "learning_rate": 0.0004114724351447873, + "loss": 1.8346, + "step": 3208 + }, + { + "epoch": 0.31337890625, + "grad_norm": 0.24078041315078735, + "learning_rate": 0.0004114158903427754, + "loss": 1.8837, + "step": 3209 + }, + { + "epoch": 0.3134765625, + "grad_norm": 0.2905030846595764, + "learning_rate": 0.00041135933191402456, + "loss": 1.8353, + "step": 3210 + }, + { + "epoch": 0.31357421875, + "grad_norm": 0.3086714446544647, + "learning_rate": 0.0004113027598641845, + "loss": 1.8614, + "step": 3211 + }, + { + "epoch": 0.313671875, + "grad_norm": 0.21693044900894165, + "learning_rate": 0.00041124617419890626, + "loss": 1.8456, + "step": 3212 + }, + { + "epoch": 0.31376953125, + "grad_norm": 0.2788260579109192, + "learning_rate": 0.0004111895749238422, + "loss": 1.8971, + "step": 3213 + }, + { + "epoch": 0.3138671875, + "grad_norm": 0.24722589552402496, + "learning_rate": 0.0004111329620446462, + "loss": 1.8583, + "step": 3214 + }, + { + "epoch": 0.31396484375, + "grad_norm": 0.19354423880577087, + "learning_rate": 0.00041107633556697326, + "loss": 1.8987, + "step": 3215 + }, + { + "epoch": 0.3140625, + "grad_norm": 0.2138962596654892, + "learning_rate": 0.00041101969549648, + "loss": 1.832, + "step": 3216 + }, + { + "epoch": 0.31416015625, + "grad_norm": 0.23529689013957977, + "learning_rate": 0.00041096304183882416, + "loss": 1.8557, + "step": 3217 + }, + { + "epoch": 0.3142578125, + "grad_norm": 0.2264302521944046, + "learning_rate": 0.000410906374599665, + "loss": 1.9008, + "step": 3218 + }, + { + "epoch": 0.31435546875, + "grad_norm": 0.27234387397766113, + "learning_rate": 0.00041084969378466294, + "loss": 1.851, + "step": 3219 + }, + { + "epoch": 0.314453125, + "grad_norm": 0.18110591173171997, + "learning_rate": 0.00041079299939948, + "loss": 1.8876, + "step": 3220 + }, + { + "epoch": 0.31455078125, + "grad_norm": 0.3096265196800232, + "learning_rate": 0.0004107362914497795, + "loss": 1.8731, + "step": 3221 + }, + { + "epoch": 0.3146484375, + "grad_norm": 0.26124435663223267, + "learning_rate": 0.000410679569941226, + "loss": 1.9025, + "step": 3222 + }, + { + "epoch": 0.31474609375, + "grad_norm": 0.2412269413471222, + "learning_rate": 0.00041062283487948545, + "loss": 1.8518, + "step": 3223 + }, + { + "epoch": 0.31484375, + "grad_norm": 0.29512709379196167, + "learning_rate": 0.0004105660862702252, + "loss": 1.8892, + "step": 3224 + }, + { + "epoch": 0.31494140625, + "grad_norm": 0.18174058198928833, + "learning_rate": 0.0004105093241191139, + "loss": 1.8556, + "step": 3225 + }, + { + "epoch": 0.3150390625, + "grad_norm": 0.30046752095222473, + "learning_rate": 0.0004104525484318217, + "loss": 1.8334, + "step": 3226 + }, + { + "epoch": 0.31513671875, + "grad_norm": 0.21368588507175446, + "learning_rate": 0.0004103957592140199, + "loss": 1.8858, + "step": 3227 + }, + { + "epoch": 0.315234375, + "grad_norm": 0.32934334874153137, + "learning_rate": 0.00041033895647138126, + "loss": 1.8602, + "step": 3228 + }, + { + "epoch": 0.31533203125, + "grad_norm": 0.25783514976501465, + "learning_rate": 0.0004102821402095798, + "loss": 1.875, + "step": 3229 + }, + { + "epoch": 0.3154296875, + "grad_norm": 0.280341774225235, + "learning_rate": 0.0004102253104342911, + "loss": 1.8705, + "step": 3230 + }, + { + "epoch": 0.31552734375, + "grad_norm": 0.25460684299468994, + "learning_rate": 0.00041016846715119184, + "loss": 1.8636, + "step": 3231 + }, + { + "epoch": 0.315625, + "grad_norm": 0.28985974192619324, + "learning_rate": 0.0004101116103659602, + "loss": 1.85, + "step": 3232 + }, + { + "epoch": 0.31572265625, + "grad_norm": 0.22553981840610504, + "learning_rate": 0.00041005474008427563, + "loss": 1.8803, + "step": 3233 + }, + { + "epoch": 0.3158203125, + "grad_norm": 0.23986418545246124, + "learning_rate": 0.0004099978563118191, + "loss": 1.8554, + "step": 3234 + }, + { + "epoch": 0.31591796875, + "grad_norm": 0.2207421511411667, + "learning_rate": 0.0004099409590542726, + "loss": 1.9089, + "step": 3235 + }, + { + "epoch": 0.316015625, + "grad_norm": 0.21956925094127655, + "learning_rate": 0.00040988404831731977, + "loss": 1.9027, + "step": 3236 + }, + { + "epoch": 0.31611328125, + "grad_norm": 0.2621534466743469, + "learning_rate": 0.00040982712410664557, + "loss": 1.9038, + "step": 3237 + }, + { + "epoch": 0.3162109375, + "grad_norm": 0.26143181324005127, + "learning_rate": 0.000409770186427936, + "loss": 1.8658, + "step": 3238 + }, + { + "epoch": 0.31630859375, + "grad_norm": 0.25201505422592163, + "learning_rate": 0.0004097132352868788, + "loss": 1.8346, + "step": 3239 + }, + { + "epoch": 0.31640625, + "grad_norm": 0.20869751274585724, + "learning_rate": 0.0004096562706891629, + "loss": 1.8839, + "step": 3240 + }, + { + "epoch": 0.31650390625, + "grad_norm": 0.26148557662963867, + "learning_rate": 0.00040959929264047855, + "loss": 1.8852, + "step": 3241 + }, + { + "epoch": 0.3166015625, + "grad_norm": 0.2597649097442627, + "learning_rate": 0.0004095423011465172, + "loss": 1.8635, + "step": 3242 + }, + { + "epoch": 0.31669921875, + "grad_norm": 0.24144643545150757, + "learning_rate": 0.0004094852962129719, + "loss": 1.8667, + "step": 3243 + }, + { + "epoch": 0.316796875, + "grad_norm": 0.2437174767255783, + "learning_rate": 0.00040942827784553703, + "loss": 1.8658, + "step": 3244 + }, + { + "epoch": 0.31689453125, + "grad_norm": 0.20527783036231995, + "learning_rate": 0.00040937124604990805, + "loss": 1.8821, + "step": 3245 + }, + { + "epoch": 0.3169921875, + "grad_norm": 0.23653388023376465, + "learning_rate": 0.00040931420083178206, + "loss": 1.8754, + "step": 3246 + }, + { + "epoch": 0.31708984375, + "grad_norm": 0.2562256455421448, + "learning_rate": 0.0004092571421968573, + "loss": 1.8726, + "step": 3247 + }, + { + "epoch": 0.3171875, + "grad_norm": 0.21176593005657196, + "learning_rate": 0.0004092000701508335, + "loss": 1.8885, + "step": 3248 + }, + { + "epoch": 0.31728515625, + "grad_norm": 0.22442927956581116, + "learning_rate": 0.0004091429846994115, + "loss": 1.8794, + "step": 3249 + }, + { + "epoch": 0.3173828125, + "grad_norm": 0.20349444448947906, + "learning_rate": 0.0004090858858482938, + "loss": 1.8958, + "step": 3250 + }, + { + "epoch": 0.31748046875, + "grad_norm": 0.2182251513004303, + "learning_rate": 0.000409028773603184, + "loss": 1.8684, + "step": 3251 + }, + { + "epoch": 0.317578125, + "grad_norm": 0.21095684170722961, + "learning_rate": 0.00040897164796978704, + "loss": 1.8564, + "step": 3252 + }, + { + "epoch": 0.31767578125, + "grad_norm": 0.2585778534412384, + "learning_rate": 0.0004089145089538094, + "loss": 1.8582, + "step": 3253 + }, + { + "epoch": 0.3177734375, + "grad_norm": 0.23634901642799377, + "learning_rate": 0.00040885735656095865, + "loss": 1.8564, + "step": 3254 + }, + { + "epoch": 0.31787109375, + "grad_norm": 0.2584446966648102, + "learning_rate": 0.0004088001907969439, + "loss": 1.9066, + "step": 3255 + }, + { + "epoch": 0.31796875, + "grad_norm": 0.24975843727588654, + "learning_rate": 0.00040874301166747535, + "loss": 1.8652, + "step": 3256 + }, + { + "epoch": 0.31806640625, + "grad_norm": 0.2328788936138153, + "learning_rate": 0.0004086858191782649, + "loss": 1.8959, + "step": 3257 + }, + { + "epoch": 0.3181640625, + "grad_norm": 0.198279470205307, + "learning_rate": 0.00040862861333502536, + "loss": 1.8361, + "step": 3258 + }, + { + "epoch": 0.31826171875, + "grad_norm": 0.21981123089790344, + "learning_rate": 0.00040857139414347127, + "loss": 1.8791, + "step": 3259 + }, + { + "epoch": 0.318359375, + "grad_norm": 0.19764897227287292, + "learning_rate": 0.0004085141616093182, + "loss": 1.8708, + "step": 3260 + }, + { + "epoch": 0.31845703125, + "grad_norm": 0.23870030045509338, + "learning_rate": 0.0004084569157382833, + "loss": 1.8774, + "step": 3261 + }, + { + "epoch": 0.3185546875, + "grad_norm": 0.23177386820316315, + "learning_rate": 0.00040839965653608477, + "loss": 1.9043, + "step": 3262 + }, + { + "epoch": 0.31865234375, + "grad_norm": 0.21266654133796692, + "learning_rate": 0.00040834238400844236, + "loss": 1.9, + "step": 3263 + }, + { + "epoch": 0.31875, + "grad_norm": 0.20720922946929932, + "learning_rate": 0.0004082850981610771, + "loss": 1.8461, + "step": 3264 + }, + { + "epoch": 0.31884765625, + "grad_norm": 0.18572098016738892, + "learning_rate": 0.00040822779899971133, + "loss": 1.8771, + "step": 3265 + }, + { + "epoch": 0.3189453125, + "grad_norm": 0.1953069120645523, + "learning_rate": 0.0004081704865300688, + "loss": 1.9015, + "step": 3266 + }, + { + "epoch": 0.31904296875, + "grad_norm": 0.19447281956672668, + "learning_rate": 0.00040811316075787434, + "loss": 1.8834, + "step": 3267 + }, + { + "epoch": 0.319140625, + "grad_norm": 0.17827379703521729, + "learning_rate": 0.00040805582168885443, + "loss": 1.7889, + "step": 3268 + }, + { + "epoch": 0.31923828125, + "grad_norm": 0.20843571424484253, + "learning_rate": 0.0004079984693287368, + "loss": 1.8655, + "step": 3269 + }, + { + "epoch": 0.3193359375, + "grad_norm": 0.1796947717666626, + "learning_rate": 0.00040794110368325023, + "loss": 1.9053, + "step": 3270 + }, + { + "epoch": 0.31943359375, + "grad_norm": 0.1886366456747055, + "learning_rate": 0.0004078837247581252, + "loss": 1.8844, + "step": 3271 + }, + { + "epoch": 0.31953125, + "grad_norm": 0.2169746607542038, + "learning_rate": 0.00040782633255909324, + "loss": 1.8976, + "step": 3272 + }, + { + "epoch": 0.31962890625, + "grad_norm": 0.1918516904115677, + "learning_rate": 0.00040776892709188745, + "loss": 1.9083, + "step": 3273 + }, + { + "epoch": 0.3197265625, + "grad_norm": 0.22316913306713104, + "learning_rate": 0.0004077115083622421, + "loss": 1.8755, + "step": 3274 + }, + { + "epoch": 0.31982421875, + "grad_norm": 0.19910778105258942, + "learning_rate": 0.0004076540763758927, + "loss": 1.8486, + "step": 3275 + }, + { + "epoch": 0.319921875, + "grad_norm": 0.23334568738937378, + "learning_rate": 0.0004075966311385764, + "loss": 1.8551, + "step": 3276 + }, + { + "epoch": 0.32001953125, + "grad_norm": 0.25610196590423584, + "learning_rate": 0.00040753917265603124, + "loss": 1.8898, + "step": 3277 + }, + { + "epoch": 0.3201171875, + "grad_norm": 0.245743066072464, + "learning_rate": 0.000407481700933997, + "loss": 1.8583, + "step": 3278 + }, + { + "epoch": 0.32021484375, + "grad_norm": 0.3178372383117676, + "learning_rate": 0.0004074242159782145, + "loss": 1.875, + "step": 3279 + }, + { + "epoch": 0.3203125, + "grad_norm": 0.2785772979259491, + "learning_rate": 0.000407366717794426, + "loss": 1.8141, + "step": 3280 + }, + { + "epoch": 0.32041015625, + "grad_norm": 0.2157028317451477, + "learning_rate": 0.000407309206388375, + "loss": 1.8298, + "step": 3281 + }, + { + "epoch": 0.3205078125, + "grad_norm": 0.28972703218460083, + "learning_rate": 0.0004072516817658065, + "loss": 1.8843, + "step": 3282 + }, + { + "epoch": 0.32060546875, + "grad_norm": 0.2404652088880539, + "learning_rate": 0.0004071941439324666, + "loss": 1.9094, + "step": 3283 + }, + { + "epoch": 0.320703125, + "grad_norm": 0.2777053117752075, + "learning_rate": 0.0004071365928941029, + "loss": 1.8456, + "step": 3284 + }, + { + "epoch": 0.32080078125, + "grad_norm": 0.2563461363315582, + "learning_rate": 0.00040707902865646405, + "loss": 1.8749, + "step": 3285 + }, + { + "epoch": 0.3208984375, + "grad_norm": 0.27938413619995117, + "learning_rate": 0.00040702145122530054, + "loss": 1.8811, + "step": 3286 + }, + { + "epoch": 0.32099609375, + "grad_norm": 0.3102744519710541, + "learning_rate": 0.00040696386060636356, + "loss": 1.8909, + "step": 3287 + }, + { + "epoch": 0.32109375, + "grad_norm": 0.23900067806243896, + "learning_rate": 0.000406906256805406, + "loss": 1.8604, + "step": 3288 + }, + { + "epoch": 0.32119140625, + "grad_norm": 0.30121737718582153, + "learning_rate": 0.000406848639828182, + "loss": 1.8719, + "step": 3289 + }, + { + "epoch": 0.3212890625, + "grad_norm": 0.25247108936309814, + "learning_rate": 0.0004067910096804469, + "loss": 1.8831, + "step": 3290 + }, + { + "epoch": 0.32138671875, + "grad_norm": 0.32374677062034607, + "learning_rate": 0.0004067333663679576, + "loss": 1.8384, + "step": 3291 + }, + { + "epoch": 0.321484375, + "grad_norm": 0.2805159091949463, + "learning_rate": 0.00040667570989647196, + "loss": 1.8153, + "step": 3292 + }, + { + "epoch": 0.32158203125, + "grad_norm": 0.1982172280550003, + "learning_rate": 0.00040661804027174945, + "loss": 1.8686, + "step": 3293 + }, + { + "epoch": 0.3216796875, + "grad_norm": 0.2745567560195923, + "learning_rate": 0.0004065603574995508, + "loss": 1.8745, + "step": 3294 + }, + { + "epoch": 0.32177734375, + "grad_norm": 0.2640945315361023, + "learning_rate": 0.000406502661585638, + "loss": 1.8627, + "step": 3295 + }, + { + "epoch": 0.321875, + "grad_norm": 0.21688511967658997, + "learning_rate": 0.00040644495253577416, + "loss": 1.8522, + "step": 3296 + }, + { + "epoch": 0.32197265625, + "grad_norm": 0.22432774305343628, + "learning_rate": 0.0004063872303557241, + "loss": 1.8475, + "step": 3297 + }, + { + "epoch": 0.3220703125, + "grad_norm": 0.27366623282432556, + "learning_rate": 0.0004063294950512538, + "loss": 1.8898, + "step": 3298 + }, + { + "epoch": 0.32216796875, + "grad_norm": 0.23350487649440765, + "learning_rate": 0.0004062717466281304, + "loss": 1.8491, + "step": 3299 + }, + { + "epoch": 0.322265625, + "grad_norm": 0.18709498643875122, + "learning_rate": 0.0004062139850921225, + "loss": 1.886, + "step": 3300 + }, + { + "epoch": 0.32236328125, + "grad_norm": 0.23382367193698883, + "learning_rate": 0.0004061562104489999, + "loss": 1.8308, + "step": 3301 + }, + { + "epoch": 0.3224609375, + "grad_norm": 0.22870516777038574, + "learning_rate": 0.00040609842270453373, + "loss": 1.8958, + "step": 3302 + }, + { + "epoch": 0.32255859375, + "grad_norm": 0.20954212546348572, + "learning_rate": 0.0004060406218644966, + "loss": 1.8669, + "step": 3303 + }, + { + "epoch": 0.32265625, + "grad_norm": 0.1893022060394287, + "learning_rate": 0.00040598280793466236, + "loss": 1.8881, + "step": 3304 + }, + { + "epoch": 0.32275390625, + "grad_norm": 0.229387104511261, + "learning_rate": 0.00040592498092080595, + "loss": 1.8944, + "step": 3305 + }, + { + "epoch": 0.3228515625, + "grad_norm": 0.22653591632843018, + "learning_rate": 0.00040586714082870386, + "loss": 1.8296, + "step": 3306 + }, + { + "epoch": 0.32294921875, + "grad_norm": 0.18969039618968964, + "learning_rate": 0.00040580928766413376, + "loss": 1.8369, + "step": 3307 + }, + { + "epoch": 0.323046875, + "grad_norm": 0.214459627866745, + "learning_rate": 0.00040575142143287464, + "loss": 1.8786, + "step": 3308 + }, + { + "epoch": 0.32314453125, + "grad_norm": 0.20712082087993622, + "learning_rate": 0.00040569354214070694, + "loss": 1.8364, + "step": 3309 + }, + { + "epoch": 0.3232421875, + "grad_norm": 0.20017264783382416, + "learning_rate": 0.00040563564979341217, + "loss": 1.8847, + "step": 3310 + }, + { + "epoch": 0.32333984375, + "grad_norm": 0.20339415967464447, + "learning_rate": 0.00040557774439677334, + "loss": 1.829, + "step": 3311 + }, + { + "epoch": 0.3234375, + "grad_norm": 0.2498069703578949, + "learning_rate": 0.00040551982595657464, + "loss": 1.8968, + "step": 3312 + }, + { + "epoch": 0.32353515625, + "grad_norm": 0.1993158906698227, + "learning_rate": 0.0004054618944786017, + "loss": 1.8516, + "step": 3313 + }, + { + "epoch": 0.3236328125, + "grad_norm": 0.21111157536506653, + "learning_rate": 0.0004054039499686412, + "loss": 1.8919, + "step": 3314 + }, + { + "epoch": 0.32373046875, + "grad_norm": 0.20167800784111023, + "learning_rate": 0.0004053459924324814, + "loss": 1.863, + "step": 3315 + }, + { + "epoch": 0.323828125, + "grad_norm": 0.26367267966270447, + "learning_rate": 0.00040528802187591174, + "loss": 1.8788, + "step": 3316 + }, + { + "epoch": 0.32392578125, + "grad_norm": 0.18443547189235687, + "learning_rate": 0.0004052300383047229, + "loss": 1.8753, + "step": 3317 + }, + { + "epoch": 0.3240234375, + "grad_norm": 0.20379985868930817, + "learning_rate": 0.000405172041724707, + "loss": 1.8347, + "step": 3318 + }, + { + "epoch": 0.32412109375, + "grad_norm": 0.1891467273235321, + "learning_rate": 0.0004051140321416574, + "loss": 1.8697, + "step": 3319 + }, + { + "epoch": 0.32421875, + "grad_norm": 0.19938014447689056, + "learning_rate": 0.0004050560095613686, + "loss": 1.9285, + "step": 3320 + }, + { + "epoch": 0.32431640625, + "grad_norm": 0.23373723030090332, + "learning_rate": 0.0004049979739896367, + "loss": 1.9126, + "step": 3321 + }, + { + "epoch": 0.3244140625, + "grad_norm": 0.199408158659935, + "learning_rate": 0.00040493992543225887, + "loss": 1.8732, + "step": 3322 + }, + { + "epoch": 0.32451171875, + "grad_norm": 0.2107379138469696, + "learning_rate": 0.00040488186389503364, + "loss": 1.8799, + "step": 3323 + }, + { + "epoch": 0.324609375, + "grad_norm": 0.1870202124118805, + "learning_rate": 0.00040482378938376084, + "loss": 1.8878, + "step": 3324 + }, + { + "epoch": 0.32470703125, + "grad_norm": 0.19757448136806488, + "learning_rate": 0.0004047657019042416, + "loss": 1.8622, + "step": 3325 + }, + { + "epoch": 0.3248046875, + "grad_norm": 0.21575550734996796, + "learning_rate": 0.0004047076014622784, + "loss": 1.82, + "step": 3326 + }, + { + "epoch": 0.32490234375, + "grad_norm": 0.2778875231742859, + "learning_rate": 0.00040464948806367496, + "loss": 1.8722, + "step": 3327 + }, + { + "epoch": 0.325, + "grad_norm": 0.27637016773223877, + "learning_rate": 0.00040459136171423624, + "loss": 1.9002, + "step": 3328 + }, + { + "epoch": 0.32509765625, + "grad_norm": 0.2065366804599762, + "learning_rate": 0.0004045332224197685, + "loss": 1.8722, + "step": 3329 + }, + { + "epoch": 0.3251953125, + "grad_norm": 0.23534739017486572, + "learning_rate": 0.00040447507018607945, + "loss": 1.8721, + "step": 3330 + }, + { + "epoch": 0.32529296875, + "grad_norm": 0.2959147095680237, + "learning_rate": 0.0004044169050189779, + "loss": 1.8888, + "step": 3331 + }, + { + "epoch": 0.325390625, + "grad_norm": 0.2515638470649719, + "learning_rate": 0.00040435872692427404, + "loss": 1.8613, + "step": 3332 + }, + { + "epoch": 0.32548828125, + "grad_norm": 0.2147361934185028, + "learning_rate": 0.00040430053590777943, + "loss": 1.8826, + "step": 3333 + }, + { + "epoch": 0.3255859375, + "grad_norm": 0.26969170570373535, + "learning_rate": 0.0004042423319753068, + "loss": 1.8766, + "step": 3334 + }, + { + "epoch": 0.32568359375, + "grad_norm": 0.20632268488407135, + "learning_rate": 0.0004041841151326702, + "loss": 1.859, + "step": 3335 + }, + { + "epoch": 0.32578125, + "grad_norm": 0.21272492408752441, + "learning_rate": 0.0004041258853856849, + "loss": 1.8988, + "step": 3336 + }, + { + "epoch": 0.32587890625, + "grad_norm": 0.18938429653644562, + "learning_rate": 0.0004040676427401676, + "loss": 1.8779, + "step": 3337 + }, + { + "epoch": 0.3259765625, + "grad_norm": 0.21298819780349731, + "learning_rate": 0.00040400938720193625, + "loss": 1.9124, + "step": 3338 + }, + { + "epoch": 0.32607421875, + "grad_norm": 0.24898596107959747, + "learning_rate": 0.00040395111877681005, + "loss": 1.8726, + "step": 3339 + }, + { + "epoch": 0.326171875, + "grad_norm": 0.2002260535955429, + "learning_rate": 0.0004038928374706095, + "loss": 1.9205, + "step": 3340 + }, + { + "epoch": 0.32626953125, + "grad_norm": 0.2754443287849426, + "learning_rate": 0.0004038345432891564, + "loss": 1.9182, + "step": 3341 + }, + { + "epoch": 0.3263671875, + "grad_norm": 0.22315667569637299, + "learning_rate": 0.0004037762362382737, + "loss": 1.8561, + "step": 3342 + }, + { + "epoch": 0.32646484375, + "grad_norm": 0.22198979556560516, + "learning_rate": 0.0004037179163237859, + "loss": 1.8672, + "step": 3343 + }, + { + "epoch": 0.3265625, + "grad_norm": 0.19733594357967377, + "learning_rate": 0.0004036595835515186, + "loss": 1.8658, + "step": 3344 + }, + { + "epoch": 0.32666015625, + "grad_norm": 0.213151216506958, + "learning_rate": 0.00040360123792729875, + "loss": 1.8732, + "step": 3345 + }, + { + "epoch": 0.3267578125, + "grad_norm": 0.19012810289859772, + "learning_rate": 0.0004035428794569545, + "loss": 1.8884, + "step": 3346 + }, + { + "epoch": 0.32685546875, + "grad_norm": 0.23143506050109863, + "learning_rate": 0.0004034845081463154, + "loss": 1.878, + "step": 3347 + }, + { + "epoch": 0.326953125, + "grad_norm": 0.26487472653388977, + "learning_rate": 0.0004034261240012122, + "loss": 1.8587, + "step": 3348 + }, + { + "epoch": 0.32705078125, + "grad_norm": 0.1884104460477829, + "learning_rate": 0.000403367727027477, + "loss": 1.9078, + "step": 3349 + }, + { + "epoch": 0.3271484375, + "grad_norm": 0.2410648763179779, + "learning_rate": 0.00040330931723094304, + "loss": 1.8787, + "step": 3350 + }, + { + "epoch": 0.32724609375, + "grad_norm": 0.20756936073303223, + "learning_rate": 0.00040325089461744503, + "loss": 1.8461, + "step": 3351 + }, + { + "epoch": 0.32734375, + "grad_norm": 0.22132349014282227, + "learning_rate": 0.0004031924591928189, + "loss": 1.8745, + "step": 3352 + }, + { + "epoch": 0.32744140625, + "grad_norm": 0.26123854517936707, + "learning_rate": 0.0004031340109629017, + "loss": 1.8589, + "step": 3353 + }, + { + "epoch": 0.3275390625, + "grad_norm": 0.29072481393814087, + "learning_rate": 0.000403075549933532, + "loss": 1.866, + "step": 3354 + }, + { + "epoch": 0.32763671875, + "grad_norm": 0.18452265858650208, + "learning_rate": 0.0004030170761105495, + "loss": 1.8764, + "step": 3355 + }, + { + "epoch": 0.327734375, + "grad_norm": 0.28800269961357117, + "learning_rate": 0.00040295858949979525, + "loss": 1.8572, + "step": 3356 + }, + { + "epoch": 0.32783203125, + "grad_norm": 0.21224476397037506, + "learning_rate": 0.00040290009010711153, + "loss": 1.8734, + "step": 3357 + }, + { + "epoch": 0.3279296875, + "grad_norm": 0.4747043251991272, + "learning_rate": 0.00040284157793834185, + "loss": 1.8632, + "step": 3358 + }, + { + "epoch": 0.32802734375, + "grad_norm": 0.2504129409790039, + "learning_rate": 0.0004027830529993312, + "loss": 1.92, + "step": 3359 + }, + { + "epoch": 0.328125, + "grad_norm": 0.25447171926498413, + "learning_rate": 0.0004027245152959256, + "loss": 1.8547, + "step": 3360 + }, + { + "epoch": 0.32822265625, + "grad_norm": 0.2585792541503906, + "learning_rate": 0.00040266596483397234, + "loss": 1.8402, + "step": 3361 + }, + { + "epoch": 0.3283203125, + "grad_norm": 0.25113922357559204, + "learning_rate": 0.00040260740161932024, + "loss": 1.8564, + "step": 3362 + }, + { + "epoch": 0.32841796875, + "grad_norm": 0.24170421063899994, + "learning_rate": 0.0004025488256578193, + "loss": 1.8804, + "step": 3363 + }, + { + "epoch": 0.328515625, + "grad_norm": 0.20581528544425964, + "learning_rate": 0.0004024902369553207, + "loss": 1.8429, + "step": 3364 + }, + { + "epoch": 0.32861328125, + "grad_norm": 0.24743464589118958, + "learning_rate": 0.0004024316355176768, + "loss": 1.8638, + "step": 3365 + }, + { + "epoch": 0.3287109375, + "grad_norm": 0.1992998570203781, + "learning_rate": 0.0004023730213507414, + "loss": 1.852, + "step": 3366 + }, + { + "epoch": 0.32880859375, + "grad_norm": 0.29643985629081726, + "learning_rate": 0.00040231439446036967, + "loss": 1.8801, + "step": 3367 + }, + { + "epoch": 0.32890625, + "grad_norm": 0.19637344777584076, + "learning_rate": 0.00040225575485241784, + "loss": 1.9142, + "step": 3368 + }, + { + "epoch": 0.32900390625, + "grad_norm": 0.3350497782230377, + "learning_rate": 0.00040219710253274344, + "loss": 1.8794, + "step": 3369 + }, + { + "epoch": 0.3291015625, + "grad_norm": 0.33154067397117615, + "learning_rate": 0.00040213843750720534, + "loss": 1.8343, + "step": 3370 + }, + { + "epoch": 0.32919921875, + "grad_norm": 0.34300824999809265, + "learning_rate": 0.00040207975978166376, + "loss": 1.8755, + "step": 3371 + }, + { + "epoch": 0.329296875, + "grad_norm": 0.21365049481391907, + "learning_rate": 0.00040202106936197994, + "loss": 1.8392, + "step": 3372 + }, + { + "epoch": 0.32939453125, + "grad_norm": 0.3048967123031616, + "learning_rate": 0.00040196236625401666, + "loss": 1.9099, + "step": 3373 + }, + { + "epoch": 0.3294921875, + "grad_norm": 0.30764639377593994, + "learning_rate": 0.00040190365046363776, + "loss": 1.865, + "step": 3374 + }, + { + "epoch": 0.32958984375, + "grad_norm": 0.2561836540699005, + "learning_rate": 0.0004018449219967084, + "loss": 1.8128, + "step": 3375 + }, + { + "epoch": 0.3296875, + "grad_norm": 0.26492762565612793, + "learning_rate": 0.0004017861808590951, + "loss": 1.8447, + "step": 3376 + }, + { + "epoch": 0.32978515625, + "grad_norm": 0.2783060073852539, + "learning_rate": 0.00040172742705666553, + "loss": 1.86, + "step": 3377 + }, + { + "epoch": 0.3298828125, + "grad_norm": 0.2328444868326187, + "learning_rate": 0.0004016686605952888, + "loss": 1.893, + "step": 3378 + }, + { + "epoch": 0.32998046875, + "grad_norm": 0.18817414343357086, + "learning_rate": 0.0004016098814808349, + "loss": 1.8713, + "step": 3379 + }, + { + "epoch": 0.330078125, + "grad_norm": 0.22875960171222687, + "learning_rate": 0.00040155108971917566, + "loss": 1.882, + "step": 3380 + }, + { + "epoch": 0.33017578125, + "grad_norm": 0.20584222674369812, + "learning_rate": 0.00040149228531618364, + "loss": 1.8385, + "step": 3381 + }, + { + "epoch": 0.3302734375, + "grad_norm": 0.22708621621131897, + "learning_rate": 0.00040143346827773297, + "loss": 1.8348, + "step": 3382 + }, + { + "epoch": 0.33037109375, + "grad_norm": 0.27723783254623413, + "learning_rate": 0.0004013746386096988, + "loss": 1.878, + "step": 3383 + }, + { + "epoch": 0.33046875, + "grad_norm": 0.2413671463727951, + "learning_rate": 0.0004013157963179579, + "loss": 1.8662, + "step": 3384 + }, + { + "epoch": 0.33056640625, + "grad_norm": 0.21263569593429565, + "learning_rate": 0.00040125694140838805, + "loss": 1.8447, + "step": 3385 + }, + { + "epoch": 0.3306640625, + "grad_norm": 0.2437567114830017, + "learning_rate": 0.0004011980738868682, + "loss": 1.8617, + "step": 3386 + }, + { + "epoch": 0.33076171875, + "grad_norm": 0.2796415686607361, + "learning_rate": 0.00040113919375927887, + "loss": 1.9222, + "step": 3387 + }, + { + "epoch": 0.330859375, + "grad_norm": 0.22173389792442322, + "learning_rate": 0.00040108030103150143, + "loss": 1.8942, + "step": 3388 + }, + { + "epoch": 0.33095703125, + "grad_norm": 0.23705333471298218, + "learning_rate": 0.000401021395709419, + "loss": 1.8621, + "step": 3389 + }, + { + "epoch": 0.3310546875, + "grad_norm": 0.2618677020072937, + "learning_rate": 0.0004009624777989156, + "loss": 1.9274, + "step": 3390 + }, + { + "epoch": 0.33115234375, + "grad_norm": 0.23887130618095398, + "learning_rate": 0.00040090354730587654, + "loss": 1.8938, + "step": 3391 + }, + { + "epoch": 0.33125, + "grad_norm": 0.2295779585838318, + "learning_rate": 0.0004008446042361885, + "loss": 1.8543, + "step": 3392 + }, + { + "epoch": 0.33134765625, + "grad_norm": 0.21946710348129272, + "learning_rate": 0.0004007856485957394, + "loss": 1.8599, + "step": 3393 + }, + { + "epoch": 0.3314453125, + "grad_norm": 0.22274813055992126, + "learning_rate": 0.00040072668039041844, + "loss": 1.8412, + "step": 3394 + }, + { + "epoch": 0.33154296875, + "grad_norm": 0.22093729674816132, + "learning_rate": 0.0004006676996261159, + "loss": 1.8838, + "step": 3395 + }, + { + "epoch": 0.331640625, + "grad_norm": 0.20294345915317535, + "learning_rate": 0.0004006087063087235, + "loss": 1.8899, + "step": 3396 + }, + { + "epoch": 0.33173828125, + "grad_norm": 0.19039151072502136, + "learning_rate": 0.00040054970044413405, + "loss": 1.9169, + "step": 3397 + }, + { + "epoch": 0.3318359375, + "grad_norm": 0.2277788668870926, + "learning_rate": 0.00040049068203824184, + "loss": 1.8882, + "step": 3398 + }, + { + "epoch": 0.33193359375, + "grad_norm": 0.22854270040988922, + "learning_rate": 0.00040043165109694223, + "loss": 1.8717, + "step": 3399 + }, + { + "epoch": 0.33203125, + "grad_norm": 0.21665067970752716, + "learning_rate": 0.00040037260762613194, + "loss": 1.8951, + "step": 3400 + }, + { + "epoch": 0.33212890625, + "grad_norm": 0.1944616436958313, + "learning_rate": 0.00040031355163170886, + "loss": 1.8353, + "step": 3401 + }, + { + "epoch": 0.3322265625, + "grad_norm": 0.20269887149333954, + "learning_rate": 0.0004002544831195721, + "loss": 1.8445, + "step": 3402 + }, + { + "epoch": 0.33232421875, + "grad_norm": 0.19559110701084137, + "learning_rate": 0.00040019540209562215, + "loss": 1.8275, + "step": 3403 + }, + { + "epoch": 0.332421875, + "grad_norm": 0.215722918510437, + "learning_rate": 0.0004001363085657606, + "loss": 1.9083, + "step": 3404 + }, + { + "epoch": 0.33251953125, + "grad_norm": 0.24516254663467407, + "learning_rate": 0.00040007720253589055, + "loss": 1.8533, + "step": 3405 + }, + { + "epoch": 0.3326171875, + "grad_norm": 0.2360994964838028, + "learning_rate": 0.000400018084011916, + "loss": 1.8852, + "step": 3406 + }, + { + "epoch": 0.33271484375, + "grad_norm": 0.20243117213249207, + "learning_rate": 0.0003999589529997424, + "loss": 1.8589, + "step": 3407 + }, + { + "epoch": 0.3328125, + "grad_norm": 0.27010393142700195, + "learning_rate": 0.00039989980950527645, + "loss": 1.8809, + "step": 3408 + }, + { + "epoch": 0.33291015625, + "grad_norm": 0.309101402759552, + "learning_rate": 0.000399840653534426, + "loss": 1.8543, + "step": 3409 + }, + { + "epoch": 0.3330078125, + "grad_norm": 0.21303214132785797, + "learning_rate": 0.00039978148509310033, + "loss": 1.8663, + "step": 3410 + }, + { + "epoch": 0.33310546875, + "grad_norm": 0.21366028487682343, + "learning_rate": 0.0003997223041872097, + "loss": 1.88, + "step": 3411 + }, + { + "epoch": 0.333203125, + "grad_norm": 0.24959337711334229, + "learning_rate": 0.0003996631108226658, + "loss": 1.8789, + "step": 3412 + }, + { + "epoch": 0.33330078125, + "grad_norm": 0.2526727616786957, + "learning_rate": 0.00039960390500538156, + "loss": 1.8924, + "step": 3413 + }, + { + "epoch": 0.3333984375, + "grad_norm": 0.1968771517276764, + "learning_rate": 0.00039954468674127116, + "loss": 1.8682, + "step": 3414 + }, + { + "epoch": 0.33349609375, + "grad_norm": 0.22142840921878815, + "learning_rate": 0.0003994854560362499, + "loss": 1.8676, + "step": 3415 + }, + { + "epoch": 0.33359375, + "grad_norm": 0.26841989159584045, + "learning_rate": 0.00039942621289623437, + "loss": 1.8776, + "step": 3416 + }, + { + "epoch": 0.33369140625, + "grad_norm": 0.19234198331832886, + "learning_rate": 0.0003993669573271426, + "loss": 1.8561, + "step": 3417 + }, + { + "epoch": 0.3337890625, + "grad_norm": 0.22837333381175995, + "learning_rate": 0.0003993076893348935, + "loss": 1.901, + "step": 3418 + }, + { + "epoch": 0.33388671875, + "grad_norm": 0.2666394114494324, + "learning_rate": 0.0003992484089254075, + "loss": 1.894, + "step": 3419 + }, + { + "epoch": 0.333984375, + "grad_norm": 0.21690601110458374, + "learning_rate": 0.0003991891161046062, + "loss": 1.8313, + "step": 3420 + }, + { + "epoch": 0.33408203125, + "grad_norm": 0.2514253556728363, + "learning_rate": 0.0003991298108784125, + "loss": 1.8744, + "step": 3421 + }, + { + "epoch": 0.3341796875, + "grad_norm": 0.21778957545757294, + "learning_rate": 0.0003990704932527502, + "loss": 1.84, + "step": 3422 + }, + { + "epoch": 0.33427734375, + "grad_norm": 0.25223058462142944, + "learning_rate": 0.00039901116323354496, + "loss": 1.8226, + "step": 3423 + }, + { + "epoch": 0.334375, + "grad_norm": 0.21998263895511627, + "learning_rate": 0.0003989518208267231, + "loss": 1.87, + "step": 3424 + }, + { + "epoch": 0.33447265625, + "grad_norm": 0.21188127994537354, + "learning_rate": 0.00039889246603821244, + "loss": 1.8484, + "step": 3425 + }, + { + "epoch": 0.3345703125, + "grad_norm": 0.18874864280223846, + "learning_rate": 0.000398833098873942, + "loss": 1.87, + "step": 3426 + }, + { + "epoch": 0.33466796875, + "grad_norm": 0.22441589832305908, + "learning_rate": 0.00039877371933984214, + "loss": 1.8692, + "step": 3427 + }, + { + "epoch": 0.334765625, + "grad_norm": 0.2002934217453003, + "learning_rate": 0.0003987143274418441, + "loss": 1.8813, + "step": 3428 + }, + { + "epoch": 0.33486328125, + "grad_norm": 0.1878109574317932, + "learning_rate": 0.0003986549231858808, + "loss": 1.8756, + "step": 3429 + }, + { + "epoch": 0.3349609375, + "grad_norm": 0.19219312071800232, + "learning_rate": 0.00039859550657788615, + "loss": 1.8489, + "step": 3430 + }, + { + "epoch": 0.33505859375, + "grad_norm": 0.2394527792930603, + "learning_rate": 0.0003985360776237955, + "loss": 1.8379, + "step": 3431 + }, + { + "epoch": 0.33515625, + "grad_norm": 0.25712698698043823, + "learning_rate": 0.00039847663632954503, + "loss": 1.888, + "step": 3432 + }, + { + "epoch": 0.33525390625, + "grad_norm": 0.20241060853004456, + "learning_rate": 0.00039841718270107246, + "loss": 1.8531, + "step": 3433 + }, + { + "epoch": 0.3353515625, + "grad_norm": 0.19888858497142792, + "learning_rate": 0.00039835771674431674, + "loss": 1.8525, + "step": 3434 + }, + { + "epoch": 0.33544921875, + "grad_norm": 0.19226659834384918, + "learning_rate": 0.00039829823846521795, + "loss": 1.8784, + "step": 3435 + }, + { + "epoch": 0.335546875, + "grad_norm": 0.22404858469963074, + "learning_rate": 0.00039823874786971747, + "loss": 1.8587, + "step": 3436 + }, + { + "epoch": 0.33564453125, + "grad_norm": 0.2008732706308365, + "learning_rate": 0.0003981792449637579, + "loss": 1.8372, + "step": 3437 + }, + { + "epoch": 0.3357421875, + "grad_norm": 0.21434251964092255, + "learning_rate": 0.00039811972975328303, + "loss": 1.886, + "step": 3438 + }, + { + "epoch": 0.33583984375, + "grad_norm": 0.20970679819583893, + "learning_rate": 0.0003980602022442379, + "loss": 1.9008, + "step": 3439 + }, + { + "epoch": 0.3359375, + "grad_norm": 0.25312310457229614, + "learning_rate": 0.00039800066244256876, + "loss": 1.8862, + "step": 3440 + }, + { + "epoch": 0.33603515625, + "grad_norm": 0.21453332901000977, + "learning_rate": 0.00039794111035422317, + "loss": 1.8772, + "step": 3441 + }, + { + "epoch": 0.3361328125, + "grad_norm": 0.22672387957572937, + "learning_rate": 0.00039788154598514977, + "loss": 1.8699, + "step": 3442 + }, + { + "epoch": 0.33623046875, + "grad_norm": 0.21547792851924896, + "learning_rate": 0.0003978219693412985, + "loss": 1.9035, + "step": 3443 + }, + { + "epoch": 0.336328125, + "grad_norm": 0.23252512514591217, + "learning_rate": 0.0003977623804286207, + "loss": 1.8595, + "step": 3444 + }, + { + "epoch": 0.33642578125, + "grad_norm": 0.29749032855033875, + "learning_rate": 0.0003977027792530687, + "loss": 1.8509, + "step": 3445 + }, + { + "epoch": 0.3365234375, + "grad_norm": 0.29424184560775757, + "learning_rate": 0.00039764316582059596, + "loss": 1.8583, + "step": 3446 + }, + { + "epoch": 0.33662109375, + "grad_norm": 0.23205117881298065, + "learning_rate": 0.00039758354013715757, + "loss": 1.8871, + "step": 3447 + }, + { + "epoch": 0.33671875, + "grad_norm": 0.23557694256305695, + "learning_rate": 0.0003975239022087095, + "loss": 1.851, + "step": 3448 + }, + { + "epoch": 0.33681640625, + "grad_norm": 0.2064058780670166, + "learning_rate": 0.00039746425204120907, + "loss": 1.866, + "step": 3449 + }, + { + "epoch": 0.3369140625, + "grad_norm": 0.19623836874961853, + "learning_rate": 0.0003974045896406148, + "loss": 1.8494, + "step": 3450 + }, + { + "epoch": 0.33701171875, + "grad_norm": 0.20967812836170197, + "learning_rate": 0.00039734491501288637, + "loss": 1.8672, + "step": 3451 + }, + { + "epoch": 0.337109375, + "grad_norm": 0.2055741846561432, + "learning_rate": 0.00039728522816398487, + "loss": 1.8739, + "step": 3452 + }, + { + "epoch": 0.33720703125, + "grad_norm": 0.1989482045173645, + "learning_rate": 0.00039722552909987244, + "loss": 1.8595, + "step": 3453 + }, + { + "epoch": 0.3373046875, + "grad_norm": 0.22619061172008514, + "learning_rate": 0.00039716581782651245, + "loss": 1.8641, + "step": 3454 + }, + { + "epoch": 0.33740234375, + "grad_norm": 0.20724697411060333, + "learning_rate": 0.00039710609434986954, + "loss": 1.866, + "step": 3455 + }, + { + "epoch": 0.3375, + "grad_norm": 0.20163816213607788, + "learning_rate": 0.0003970463586759095, + "loss": 1.8354, + "step": 3456 + }, + { + "epoch": 0.33759765625, + "grad_norm": 0.22491896152496338, + "learning_rate": 0.00039698661081059963, + "loss": 1.8648, + "step": 3457 + }, + { + "epoch": 0.3376953125, + "grad_norm": 0.19193711876869202, + "learning_rate": 0.0003969268507599079, + "loss": 1.8797, + "step": 3458 + }, + { + "epoch": 0.33779296875, + "grad_norm": 0.2137691229581833, + "learning_rate": 0.00039686707852980405, + "loss": 1.862, + "step": 3459 + }, + { + "epoch": 0.337890625, + "grad_norm": 0.2840065062046051, + "learning_rate": 0.00039680729412625867, + "loss": 1.8528, + "step": 3460 + }, + { + "epoch": 0.33798828125, + "grad_norm": 0.22890503704547882, + "learning_rate": 0.0003967474975552436, + "loss": 1.8681, + "step": 3461 + }, + { + "epoch": 0.3380859375, + "grad_norm": 0.20640310645103455, + "learning_rate": 0.0003966876888227323, + "loss": 1.8319, + "step": 3462 + }, + { + "epoch": 0.33818359375, + "grad_norm": 0.20327606797218323, + "learning_rate": 0.00039662786793469887, + "loss": 1.8788, + "step": 3463 + }, + { + "epoch": 0.33828125, + "grad_norm": 0.26422062516212463, + "learning_rate": 0.0003965680348971189, + "loss": 1.8525, + "step": 3464 + }, + { + "epoch": 0.33837890625, + "grad_norm": 0.23500823974609375, + "learning_rate": 0.0003965081897159693, + "loss": 1.8777, + "step": 3465 + }, + { + "epoch": 0.3384765625, + "grad_norm": 0.19631804525852203, + "learning_rate": 0.000396448332397228, + "loss": 1.8702, + "step": 3466 + }, + { + "epoch": 0.33857421875, + "grad_norm": 0.21247118711471558, + "learning_rate": 0.0003963884629468742, + "loss": 1.8706, + "step": 3467 + }, + { + "epoch": 0.338671875, + "grad_norm": 0.23746398091316223, + "learning_rate": 0.0003963285813708884, + "loss": 1.8526, + "step": 3468 + }, + { + "epoch": 0.33876953125, + "grad_norm": 0.2166304886341095, + "learning_rate": 0.00039626868767525227, + "loss": 1.8407, + "step": 3469 + }, + { + "epoch": 0.3388671875, + "grad_norm": 0.18043971061706543, + "learning_rate": 0.00039620878186594845, + "loss": 1.8535, + "step": 3470 + }, + { + "epoch": 0.33896484375, + "grad_norm": 0.20053336024284363, + "learning_rate": 0.0003961488639489612, + "loss": 1.7842, + "step": 3471 + }, + { + "epoch": 0.3390625, + "grad_norm": 0.22193653881549835, + "learning_rate": 0.0003960889339302756, + "loss": 1.8312, + "step": 3472 + }, + { + "epoch": 0.33916015625, + "grad_norm": 0.20127463340759277, + "learning_rate": 0.00039602899181587843, + "loss": 1.9094, + "step": 3473 + }, + { + "epoch": 0.3392578125, + "grad_norm": 0.23692427575588226, + "learning_rate": 0.0003959690376117571, + "loss": 1.9139, + "step": 3474 + }, + { + "epoch": 0.33935546875, + "grad_norm": 0.23256926238536835, + "learning_rate": 0.0003959090713239006, + "loss": 1.8432, + "step": 3475 + }, + { + "epoch": 0.339453125, + "grad_norm": 0.1989685297012329, + "learning_rate": 0.00039584909295829914, + "loss": 1.8473, + "step": 3476 + }, + { + "epoch": 0.33955078125, + "grad_norm": 0.2824874222278595, + "learning_rate": 0.0003957891025209438, + "loss": 1.9134, + "step": 3477 + }, + { + "epoch": 0.3396484375, + "grad_norm": 0.293108731508255, + "learning_rate": 0.0003957291000178273, + "loss": 1.8542, + "step": 3478 + }, + { + "epoch": 0.33974609375, + "grad_norm": 0.26237407326698303, + "learning_rate": 0.00039566908545494317, + "loss": 1.8847, + "step": 3479 + }, + { + "epoch": 0.33984375, + "grad_norm": 0.2476871907711029, + "learning_rate": 0.0003956090588382865, + "loss": 1.8712, + "step": 3480 + }, + { + "epoch": 0.33994140625, + "grad_norm": 0.29409369826316833, + "learning_rate": 0.00039554902017385334, + "loss": 1.8622, + "step": 3481 + }, + { + "epoch": 0.3400390625, + "grad_norm": 0.24744457006454468, + "learning_rate": 0.0003954889694676411, + "loss": 1.8819, + "step": 3482 + }, + { + "epoch": 0.34013671875, + "grad_norm": 0.2826077342033386, + "learning_rate": 0.00039542890672564815, + "loss": 1.8725, + "step": 3483 + }, + { + "epoch": 0.340234375, + "grad_norm": 0.2639237940311432, + "learning_rate": 0.00039536883195387434, + "loss": 1.8544, + "step": 3484 + }, + { + "epoch": 0.34033203125, + "grad_norm": 0.20325662195682526, + "learning_rate": 0.0003953087451583206, + "loss": 1.8691, + "step": 3485 + }, + { + "epoch": 0.3404296875, + "grad_norm": 0.23570391535758972, + "learning_rate": 0.00039524864634498904, + "loss": 1.8841, + "step": 3486 + }, + { + "epoch": 0.34052734375, + "grad_norm": 0.2665724754333496, + "learning_rate": 0.00039518853551988303, + "loss": 1.8977, + "step": 3487 + }, + { + "epoch": 0.340625, + "grad_norm": 0.22694428265094757, + "learning_rate": 0.0003951284126890071, + "loss": 1.89, + "step": 3488 + }, + { + "epoch": 0.34072265625, + "grad_norm": 0.20151232182979584, + "learning_rate": 0.000395068277858367, + "loss": 1.8981, + "step": 3489 + }, + { + "epoch": 0.3408203125, + "grad_norm": 0.25018495321273804, + "learning_rate": 0.0003950081310339697, + "loss": 1.8307, + "step": 3490 + }, + { + "epoch": 0.34091796875, + "grad_norm": 0.22320601344108582, + "learning_rate": 0.00039494797222182316, + "loss": 1.8672, + "step": 3491 + }, + { + "epoch": 0.341015625, + "grad_norm": 0.23789243400096893, + "learning_rate": 0.00039488780142793694, + "loss": 1.8801, + "step": 3492 + }, + { + "epoch": 0.34111328125, + "grad_norm": 0.21100810170173645, + "learning_rate": 0.0003948276186583214, + "loss": 1.8492, + "step": 3493 + }, + { + "epoch": 0.3412109375, + "grad_norm": 0.20523199439048767, + "learning_rate": 0.00039476742391898837, + "loss": 1.901, + "step": 3494 + }, + { + "epoch": 0.34130859375, + "grad_norm": 0.20760668814182281, + "learning_rate": 0.0003947072172159507, + "loss": 1.8568, + "step": 3495 + }, + { + "epoch": 0.34140625, + "grad_norm": 0.25376006960868835, + "learning_rate": 0.0003946469985552226, + "loss": 1.8542, + "step": 3496 + }, + { + "epoch": 0.34150390625, + "grad_norm": 0.20873723924160004, + "learning_rate": 0.0003945867679428193, + "loss": 1.8735, + "step": 3497 + }, + { + "epoch": 0.3416015625, + "grad_norm": 0.20614324510097504, + "learning_rate": 0.0003945265253847573, + "loss": 1.8462, + "step": 3498 + }, + { + "epoch": 0.34169921875, + "grad_norm": 0.22653350234031677, + "learning_rate": 0.0003944662708870544, + "loss": 1.8914, + "step": 3499 + }, + { + "epoch": 0.341796875, + "grad_norm": 0.18062424659729004, + "learning_rate": 0.0003944060044557293, + "loss": 1.8581, + "step": 3500 + }, + { + "epoch": 0.34189453125, + "grad_norm": 0.22148776054382324, + "learning_rate": 0.00039434572609680225, + "loss": 1.8843, + "step": 3501 + }, + { + "epoch": 0.3419921875, + "grad_norm": 0.19227011501789093, + "learning_rate": 0.0003942854358162945, + "loss": 1.8677, + "step": 3502 + }, + { + "epoch": 0.34208984375, + "grad_norm": 0.20412443578243256, + "learning_rate": 0.00039422513362022844, + "loss": 1.835, + "step": 3503 + }, + { + "epoch": 0.3421875, + "grad_norm": 0.19800935685634613, + "learning_rate": 0.0003941648195146278, + "loss": 1.9221, + "step": 3504 + }, + { + "epoch": 0.34228515625, + "grad_norm": 0.2127007693052292, + "learning_rate": 0.00039410449350551737, + "loss": 1.8946, + "step": 3505 + }, + { + "epoch": 0.3423828125, + "grad_norm": 0.22856412827968597, + "learning_rate": 0.00039404415559892326, + "loss": 1.8817, + "step": 3506 + }, + { + "epoch": 0.34248046875, + "grad_norm": 0.18088886141777039, + "learning_rate": 0.0003939838058008726, + "loss": 1.8779, + "step": 3507 + }, + { + "epoch": 0.342578125, + "grad_norm": 0.19799447059631348, + "learning_rate": 0.0003939234441173938, + "loss": 1.8329, + "step": 3508 + }, + { + "epoch": 0.34267578125, + "grad_norm": 0.205015167593956, + "learning_rate": 0.00039386307055451654, + "loss": 1.9022, + "step": 3509 + }, + { + "epoch": 0.3427734375, + "grad_norm": 0.19568204879760742, + "learning_rate": 0.0003938026851182716, + "loss": 1.8265, + "step": 3510 + }, + { + "epoch": 0.34287109375, + "grad_norm": 0.1977141946554184, + "learning_rate": 0.0003937422878146909, + "loss": 1.8974, + "step": 3511 + }, + { + "epoch": 0.34296875, + "grad_norm": 0.24509556591510773, + "learning_rate": 0.0003936818786498076, + "loss": 1.8637, + "step": 3512 + }, + { + "epoch": 0.34306640625, + "grad_norm": 0.18602833151817322, + "learning_rate": 0.000393621457629656, + "loss": 1.8834, + "step": 3513 + }, + { + "epoch": 0.3431640625, + "grad_norm": 0.25452902913093567, + "learning_rate": 0.00039356102476027175, + "loss": 1.7987, + "step": 3514 + }, + { + "epoch": 0.34326171875, + "grad_norm": 0.2689136266708374, + "learning_rate": 0.0003935005800476914, + "loss": 1.8608, + "step": 3515 + }, + { + "epoch": 0.343359375, + "grad_norm": 0.17565156519412994, + "learning_rate": 0.000393440123497953, + "loss": 1.8781, + "step": 3516 + }, + { + "epoch": 0.34345703125, + "grad_norm": 0.2498348355293274, + "learning_rate": 0.0003933796551170955, + "loss": 1.8346, + "step": 3517 + }, + { + "epoch": 0.3435546875, + "grad_norm": 0.2310560792684555, + "learning_rate": 0.00039331917491115933, + "loss": 1.8462, + "step": 3518 + }, + { + "epoch": 0.34365234375, + "grad_norm": 0.24668028950691223, + "learning_rate": 0.00039325868288618565, + "loss": 1.9157, + "step": 3519 + }, + { + "epoch": 0.34375, + "grad_norm": 0.20355354249477386, + "learning_rate": 0.0003931981790482172, + "loss": 1.9128, + "step": 3520 + }, + { + "epoch": 0.34384765625, + "grad_norm": 0.25293999910354614, + "learning_rate": 0.00039313766340329795, + "loss": 1.8329, + "step": 3521 + }, + { + "epoch": 0.3439453125, + "grad_norm": 0.22218835353851318, + "learning_rate": 0.00039307713595747263, + "loss": 1.8865, + "step": 3522 + }, + { + "epoch": 0.34404296875, + "grad_norm": 0.22433200478553772, + "learning_rate": 0.0003930165967167876, + "loss": 1.8778, + "step": 3523 + }, + { + "epoch": 0.344140625, + "grad_norm": 0.22571197152137756, + "learning_rate": 0.0003929560456872899, + "loss": 1.8835, + "step": 3524 + }, + { + "epoch": 0.34423828125, + "grad_norm": 0.2275848686695099, + "learning_rate": 0.00039289548287502847, + "loss": 1.8552, + "step": 3525 + }, + { + "epoch": 0.3443359375, + "grad_norm": 0.19883055984973907, + "learning_rate": 0.00039283490828605254, + "loss": 1.8562, + "step": 3526 + }, + { + "epoch": 0.34443359375, + "grad_norm": 0.2009442299604416, + "learning_rate": 0.00039277432192641335, + "loss": 1.8354, + "step": 3527 + }, + { + "epoch": 0.34453125, + "grad_norm": 0.20756255090236664, + "learning_rate": 0.0003927137238021627, + "loss": 1.8615, + "step": 3528 + }, + { + "epoch": 0.34462890625, + "grad_norm": 0.21216446161270142, + "learning_rate": 0.00039265311391935395, + "loss": 1.9002, + "step": 3529 + }, + { + "epoch": 0.3447265625, + "grad_norm": 0.21828250586986542, + "learning_rate": 0.00039259249228404137, + "loss": 1.9665, + "step": 3530 + }, + { + "epoch": 0.34482421875, + "grad_norm": 0.22328948974609375, + "learning_rate": 0.0003925318589022806, + "loss": 1.8696, + "step": 3531 + }, + { + "epoch": 0.344921875, + "grad_norm": 0.27527397871017456, + "learning_rate": 0.00039247121378012845, + "loss": 1.8477, + "step": 3532 + }, + { + "epoch": 0.34501953125, + "grad_norm": 0.3070453703403473, + "learning_rate": 0.00039241055692364256, + "loss": 1.8264, + "step": 3533 + }, + { + "epoch": 0.3451171875, + "grad_norm": 0.22435393929481506, + "learning_rate": 0.0003923498883388824, + "loss": 1.8824, + "step": 3534 + }, + { + "epoch": 0.34521484375, + "grad_norm": 0.25616350769996643, + "learning_rate": 0.00039228920803190793, + "loss": 1.8645, + "step": 3535 + }, + { + "epoch": 0.3453125, + "grad_norm": 0.3189918100833893, + "learning_rate": 0.00039222851600878073, + "loss": 1.843, + "step": 3536 + }, + { + "epoch": 0.34541015625, + "grad_norm": 0.22394324839115143, + "learning_rate": 0.0003921678122755632, + "loss": 1.8536, + "step": 3537 + }, + { + "epoch": 0.3455078125, + "grad_norm": 0.22344903647899628, + "learning_rate": 0.00039210709683831936, + "loss": 1.8434, + "step": 3538 + }, + { + "epoch": 0.34560546875, + "grad_norm": 0.18886572122573853, + "learning_rate": 0.000392046369703114, + "loss": 1.8473, + "step": 3539 + }, + { + "epoch": 0.345703125, + "grad_norm": 0.24359601736068726, + "learning_rate": 0.00039198563087601313, + "loss": 1.8687, + "step": 3540 + }, + { + "epoch": 0.34580078125, + "grad_norm": 0.1771291345357895, + "learning_rate": 0.0003919248803630843, + "loss": 1.9114, + "step": 3541 + }, + { + "epoch": 0.3458984375, + "grad_norm": 0.25192058086395264, + "learning_rate": 0.0003918641181703957, + "loss": 1.8616, + "step": 3542 + }, + { + "epoch": 0.34599609375, + "grad_norm": 0.21408501267433167, + "learning_rate": 0.000391803344304017, + "loss": 1.8387, + "step": 3543 + }, + { + "epoch": 0.34609375, + "grad_norm": 0.2607284486293793, + "learning_rate": 0.000391742558770019, + "loss": 1.8529, + "step": 3544 + }, + { + "epoch": 0.34619140625, + "grad_norm": 0.2747364044189453, + "learning_rate": 0.0003916817615744737, + "loss": 1.8514, + "step": 3545 + }, + { + "epoch": 0.3462890625, + "grad_norm": 0.2552090585231781, + "learning_rate": 0.000391620952723454, + "loss": 1.8858, + "step": 3546 + }, + { + "epoch": 0.34638671875, + "grad_norm": 0.27006518840789795, + "learning_rate": 0.00039156013222303433, + "loss": 1.8749, + "step": 3547 + }, + { + "epoch": 0.346484375, + "grad_norm": 0.2662247121334076, + "learning_rate": 0.00039149930007929015, + "loss": 1.8688, + "step": 3548 + }, + { + "epoch": 0.34658203125, + "grad_norm": 0.2671772539615631, + "learning_rate": 0.00039143845629829795, + "loss": 1.852, + "step": 3549 + }, + { + "epoch": 0.3466796875, + "grad_norm": 0.204731747508049, + "learning_rate": 0.0003913776008861355, + "loss": 1.8848, + "step": 3550 + }, + { + "epoch": 0.34677734375, + "grad_norm": 0.27976933121681213, + "learning_rate": 0.00039131673384888173, + "loss": 1.8698, + "step": 3551 + }, + { + "epoch": 0.346875, + "grad_norm": 0.19963674247264862, + "learning_rate": 0.0003912558551926168, + "loss": 1.9433, + "step": 3552 + }, + { + "epoch": 0.34697265625, + "grad_norm": 0.2379409819841385, + "learning_rate": 0.00039119496492342175, + "loss": 1.8777, + "step": 3553 + }, + { + "epoch": 0.3470703125, + "grad_norm": 0.21821127831935883, + "learning_rate": 0.0003911340630473791, + "loss": 1.8634, + "step": 3554 + }, + { + "epoch": 0.34716796875, + "grad_norm": 0.26188230514526367, + "learning_rate": 0.0003910731495705725, + "loss": 1.8684, + "step": 3555 + }, + { + "epoch": 0.347265625, + "grad_norm": 0.2163819670677185, + "learning_rate": 0.0003910122244990866, + "loss": 1.8762, + "step": 3556 + }, + { + "epoch": 0.34736328125, + "grad_norm": 0.2630750834941864, + "learning_rate": 0.0003909512878390072, + "loss": 1.8506, + "step": 3557 + }, + { + "epoch": 0.3474609375, + "grad_norm": 0.229984313249588, + "learning_rate": 0.0003908903395964213, + "loss": 1.8492, + "step": 3558 + }, + { + "epoch": 0.34755859375, + "grad_norm": 0.22928504645824432, + "learning_rate": 0.0003908293797774172, + "loss": 1.8541, + "step": 3559 + }, + { + "epoch": 0.34765625, + "grad_norm": 0.22383026778697968, + "learning_rate": 0.0003907684083880843, + "loss": 1.8417, + "step": 3560 + }, + { + "epoch": 0.34775390625, + "grad_norm": 0.23472000658512115, + "learning_rate": 0.0003907074254345129, + "loss": 1.8546, + "step": 3561 + }, + { + "epoch": 0.3478515625, + "grad_norm": 0.20112402737140656, + "learning_rate": 0.00039064643092279487, + "loss": 1.8371, + "step": 3562 + }, + { + "epoch": 0.34794921875, + "grad_norm": 0.23486244678497314, + "learning_rate": 0.0003905854248590229, + "loss": 1.8562, + "step": 3563 + }, + { + "epoch": 0.348046875, + "grad_norm": 0.17908360064029694, + "learning_rate": 0.000390524407249291, + "loss": 1.828, + "step": 3564 + }, + { + "epoch": 0.34814453125, + "grad_norm": 0.2056552618741989, + "learning_rate": 0.00039046337809969415, + "loss": 1.8765, + "step": 3565 + }, + { + "epoch": 0.3482421875, + "grad_norm": 0.19951866567134857, + "learning_rate": 0.0003904023374163289, + "loss": 1.8921, + "step": 3566 + }, + { + "epoch": 0.34833984375, + "grad_norm": 0.22525285184383392, + "learning_rate": 0.0003903412852052925, + "loss": 1.8943, + "step": 3567 + }, + { + "epoch": 0.3484375, + "grad_norm": 0.27338072657585144, + "learning_rate": 0.00039028022147268346, + "loss": 1.8677, + "step": 3568 + }, + { + "epoch": 0.34853515625, + "grad_norm": 0.287302166223526, + "learning_rate": 0.00039021914622460157, + "loss": 1.8679, + "step": 3569 + }, + { + "epoch": 0.3486328125, + "grad_norm": 0.1998988389968872, + "learning_rate": 0.0003901580594671478, + "loss": 1.8138, + "step": 3570 + }, + { + "epoch": 0.34873046875, + "grad_norm": 0.24139219522476196, + "learning_rate": 0.00039009696120642407, + "loss": 1.8438, + "step": 3571 + }, + { + "epoch": 0.348828125, + "grad_norm": 0.21555127203464508, + "learning_rate": 0.00039003585144853353, + "loss": 1.8697, + "step": 3572 + }, + { + "epoch": 0.34892578125, + "grad_norm": 0.20976285636425018, + "learning_rate": 0.0003899747301995806, + "loss": 1.8486, + "step": 3573 + }, + { + "epoch": 0.3490234375, + "grad_norm": 0.22705312073230743, + "learning_rate": 0.0003899135974656708, + "loss": 1.8197, + "step": 3574 + }, + { + "epoch": 0.34912109375, + "grad_norm": 0.21448858082294464, + "learning_rate": 0.0003898524532529105, + "loss": 1.8833, + "step": 3575 + }, + { + "epoch": 0.34921875, + "grad_norm": 0.1872919797897339, + "learning_rate": 0.00038979129756740774, + "loss": 1.8701, + "step": 3576 + }, + { + "epoch": 0.34931640625, + "grad_norm": 0.25819647312164307, + "learning_rate": 0.00038973013041527127, + "loss": 1.8932, + "step": 3577 + }, + { + "epoch": 0.3494140625, + "grad_norm": 0.19156448543071747, + "learning_rate": 0.0003896689518026112, + "loss": 1.856, + "step": 3578 + }, + { + "epoch": 0.34951171875, + "grad_norm": 0.20140022039413452, + "learning_rate": 0.00038960776173553867, + "loss": 1.8327, + "step": 3579 + }, + { + "epoch": 0.349609375, + "grad_norm": 0.18098512291908264, + "learning_rate": 0.0003895465602201661, + "loss": 1.8791, + "step": 3580 + }, + { + "epoch": 0.34970703125, + "grad_norm": 0.20784790813922882, + "learning_rate": 0.000389485347262607, + "loss": 1.8375, + "step": 3581 + }, + { + "epoch": 0.3498046875, + "grad_norm": 0.22096554934978485, + "learning_rate": 0.0003894241228689759, + "loss": 1.8553, + "step": 3582 + }, + { + "epoch": 0.34990234375, + "grad_norm": 0.24984371662139893, + "learning_rate": 0.0003893628870453886, + "loss": 1.8573, + "step": 3583 + }, + { + "epoch": 0.35, + "grad_norm": 0.19057220220565796, + "learning_rate": 0.0003893016397979621, + "loss": 1.8712, + "step": 3584 + }, + { + "epoch": 0.35009765625, + "grad_norm": 0.22978176176548004, + "learning_rate": 0.0003892403811328144, + "loss": 1.8274, + "step": 3585 + }, + { + "epoch": 0.3501953125, + "grad_norm": 0.21941888332366943, + "learning_rate": 0.00038917911105606476, + "loss": 1.8153, + "step": 3586 + }, + { + "epoch": 0.35029296875, + "grad_norm": 0.2456945776939392, + "learning_rate": 0.0003891178295738335, + "loss": 1.8303, + "step": 3587 + }, + { + "epoch": 0.350390625, + "grad_norm": 0.2776344120502472, + "learning_rate": 0.00038905653669224185, + "loss": 1.8185, + "step": 3588 + }, + { + "epoch": 0.35048828125, + "grad_norm": 0.2225499004125595, + "learning_rate": 0.00038899523241741273, + "loss": 1.8456, + "step": 3589 + }, + { + "epoch": 0.3505859375, + "grad_norm": 0.2002578228712082, + "learning_rate": 0.00038893391675546993, + "loss": 1.8501, + "step": 3590 + }, + { + "epoch": 0.35068359375, + "grad_norm": 0.21190977096557617, + "learning_rate": 0.00038887258971253804, + "loss": 1.8311, + "step": 3591 + }, + { + "epoch": 0.35078125, + "grad_norm": 0.20795829594135284, + "learning_rate": 0.00038881125129474334, + "loss": 1.8079, + "step": 3592 + }, + { + "epoch": 0.35087890625, + "grad_norm": 0.20573373138904572, + "learning_rate": 0.0003887499015082129, + "loss": 1.9012, + "step": 3593 + }, + { + "epoch": 0.3509765625, + "grad_norm": 0.24483314156532288, + "learning_rate": 0.000388688540359075, + "loss": 1.841, + "step": 3594 + }, + { + "epoch": 0.35107421875, + "grad_norm": 0.2204584926366806, + "learning_rate": 0.00038862716785345916, + "loss": 1.834, + "step": 3595 + }, + { + "epoch": 0.351171875, + "grad_norm": 0.18937397003173828, + "learning_rate": 0.00038856578399749583, + "loss": 1.8726, + "step": 3596 + }, + { + "epoch": 0.35126953125, + "grad_norm": 0.23919473588466644, + "learning_rate": 0.0003885043887973168, + "loss": 1.8745, + "step": 3597 + }, + { + "epoch": 0.3513671875, + "grad_norm": 0.19273193180561066, + "learning_rate": 0.000388442982259055, + "loss": 1.833, + "step": 3598 + }, + { + "epoch": 0.35146484375, + "grad_norm": 0.2513469159603119, + "learning_rate": 0.0003883815643888441, + "loss": 1.8519, + "step": 3599 + }, + { + "epoch": 0.3515625, + "grad_norm": 0.3091588318347931, + "learning_rate": 0.0003883201351928195, + "loss": 1.8621, + "step": 3600 + }, + { + "epoch": 0.35166015625, + "grad_norm": 0.22832126915454865, + "learning_rate": 0.0003882586946771173, + "loss": 1.8215, + "step": 3601 + }, + { + "epoch": 0.3517578125, + "grad_norm": 0.20138737559318542, + "learning_rate": 0.00038819724284787496, + "loss": 1.8488, + "step": 3602 + }, + { + "epoch": 0.35185546875, + "grad_norm": 0.2468486875295639, + "learning_rate": 0.00038813577971123076, + "loss": 1.8553, + "step": 3603 + }, + { + "epoch": 0.351953125, + "grad_norm": 0.2534483075141907, + "learning_rate": 0.0003880743052733246, + "loss": 1.8304, + "step": 3604 + }, + { + "epoch": 0.35205078125, + "grad_norm": 0.17403554916381836, + "learning_rate": 0.000388012819540297, + "loss": 1.82, + "step": 3605 + }, + { + "epoch": 0.3521484375, + "grad_norm": 0.23867322504520416, + "learning_rate": 0.00038795132251828995, + "loss": 1.8401, + "step": 3606 + }, + { + "epoch": 0.35224609375, + "grad_norm": 0.3085857033729553, + "learning_rate": 0.00038788981421344643, + "loss": 1.8749, + "step": 3607 + }, + { + "epoch": 0.35234375, + "grad_norm": 0.24236635863780975, + "learning_rate": 0.00038782829463191056, + "loss": 1.8541, + "step": 3608 + }, + { + "epoch": 0.35244140625, + "grad_norm": 0.16068558394908905, + "learning_rate": 0.00038776676377982763, + "loss": 1.8867, + "step": 3609 + }, + { + "epoch": 0.3525390625, + "grad_norm": 0.2600993514060974, + "learning_rate": 0.00038770522166334405, + "loss": 1.872, + "step": 3610 + }, + { + "epoch": 0.35263671875, + "grad_norm": 0.21148955821990967, + "learning_rate": 0.00038764366828860736, + "loss": 1.8504, + "step": 3611 + }, + { + "epoch": 0.352734375, + "grad_norm": 0.19798903167247772, + "learning_rate": 0.00038758210366176605, + "loss": 1.8641, + "step": 3612 + }, + { + "epoch": 0.35283203125, + "grad_norm": 0.24468740820884705, + "learning_rate": 0.00038752052778896995, + "loss": 1.8666, + "step": 3613 + }, + { + "epoch": 0.3529296875, + "grad_norm": 0.2498670220375061, + "learning_rate": 0.00038745894067637005, + "loss": 1.894, + "step": 3614 + }, + { + "epoch": 0.35302734375, + "grad_norm": 0.192293182015419, + "learning_rate": 0.0003873973423301183, + "loss": 1.8973, + "step": 3615 + }, + { + "epoch": 0.353125, + "grad_norm": 0.23599712550640106, + "learning_rate": 0.00038733573275636767, + "loss": 1.8605, + "step": 3616 + }, + { + "epoch": 0.35322265625, + "grad_norm": 0.24024492502212524, + "learning_rate": 0.0003872741119612726, + "loss": 1.8377, + "step": 3617 + }, + { + "epoch": 0.3533203125, + "grad_norm": 0.18636207282543182, + "learning_rate": 0.00038721247995098846, + "loss": 1.8966, + "step": 3618 + }, + { + "epoch": 0.35341796875, + "grad_norm": 0.2846168577671051, + "learning_rate": 0.0003871508367316717, + "loss": 1.887, + "step": 3619 + }, + { + "epoch": 0.353515625, + "grad_norm": 0.2541621923446655, + "learning_rate": 0.0003870891823094798, + "loss": 1.864, + "step": 3620 + }, + { + "epoch": 0.35361328125, + "grad_norm": 0.2141297310590744, + "learning_rate": 0.0003870275166905717, + "loss": 1.8547, + "step": 3621 + }, + { + "epoch": 0.3537109375, + "grad_norm": 0.22027063369750977, + "learning_rate": 0.00038696583988110717, + "loss": 1.881, + "step": 3622 + }, + { + "epoch": 0.35380859375, + "grad_norm": 0.18557903170585632, + "learning_rate": 0.0003869041518872471, + "loss": 1.8469, + "step": 3623 + }, + { + "epoch": 0.35390625, + "grad_norm": 0.202976256608963, + "learning_rate": 0.00038684245271515365, + "loss": 1.8438, + "step": 3624 + }, + { + "epoch": 0.35400390625, + "grad_norm": 0.2126491665840149, + "learning_rate": 0.00038678074237099, + "loss": 1.8884, + "step": 3625 + }, + { + "epoch": 0.3541015625, + "grad_norm": 0.2188192456960678, + "learning_rate": 0.0003867190208609205, + "loss": 1.8504, + "step": 3626 + }, + { + "epoch": 0.35419921875, + "grad_norm": 0.2543421983718872, + "learning_rate": 0.00038665728819111056, + "loss": 1.8995, + "step": 3627 + }, + { + "epoch": 0.354296875, + "grad_norm": 0.2173098921775818, + "learning_rate": 0.0003865955443677267, + "loss": 1.9028, + "step": 3628 + }, + { + "epoch": 0.35439453125, + "grad_norm": 0.21151049435138702, + "learning_rate": 0.0003865337893969367, + "loss": 1.8721, + "step": 3629 + }, + { + "epoch": 0.3544921875, + "grad_norm": 0.24614599347114563, + "learning_rate": 0.0003864720232849091, + "loss": 1.8269, + "step": 3630 + }, + { + "epoch": 0.35458984375, + "grad_norm": 0.19173894822597504, + "learning_rate": 0.000386410246037814, + "loss": 1.9056, + "step": 3631 + }, + { + "epoch": 0.3546875, + "grad_norm": 0.2210468053817749, + "learning_rate": 0.00038634845766182235, + "loss": 1.842, + "step": 3632 + }, + { + "epoch": 0.35478515625, + "grad_norm": 0.23308482766151428, + "learning_rate": 0.00038628665816310616, + "loss": 1.8237, + "step": 3633 + }, + { + "epoch": 0.3548828125, + "grad_norm": 0.23058968782424927, + "learning_rate": 0.0003862248475478388, + "loss": 1.8754, + "step": 3634 + }, + { + "epoch": 0.35498046875, + "grad_norm": 0.2620023488998413, + "learning_rate": 0.0003861630258221945, + "loss": 1.873, + "step": 3635 + }, + { + "epoch": 0.355078125, + "grad_norm": 0.24256622791290283, + "learning_rate": 0.00038610119299234874, + "loss": 1.8666, + "step": 3636 + }, + { + "epoch": 0.35517578125, + "grad_norm": 0.23211470246315002, + "learning_rate": 0.0003860393490644781, + "loss": 1.8632, + "step": 3637 + }, + { + "epoch": 0.3552734375, + "grad_norm": 0.22089043259620667, + "learning_rate": 0.00038597749404476015, + "loss": 1.8948, + "step": 3638 + }, + { + "epoch": 0.35537109375, + "grad_norm": 0.20722050964832306, + "learning_rate": 0.00038591562793937375, + "loss": 1.8426, + "step": 3639 + }, + { + "epoch": 0.35546875, + "grad_norm": 0.2407168745994568, + "learning_rate": 0.00038585375075449874, + "loss": 1.8757, + "step": 3640 + }, + { + "epoch": 0.35556640625, + "grad_norm": 0.21094001829624176, + "learning_rate": 0.0003857918624963161, + "loss": 1.8471, + "step": 3641 + }, + { + "epoch": 0.3556640625, + "grad_norm": 0.21768639981746674, + "learning_rate": 0.00038572996317100787, + "loss": 1.8501, + "step": 3642 + }, + { + "epoch": 0.35576171875, + "grad_norm": 0.198587104678154, + "learning_rate": 0.0003856680527847574, + "loss": 1.8872, + "step": 3643 + }, + { + "epoch": 0.355859375, + "grad_norm": 0.2296200841665268, + "learning_rate": 0.0003856061313437489, + "loss": 1.8713, + "step": 3644 + }, + { + "epoch": 0.35595703125, + "grad_norm": 0.19478590786457062, + "learning_rate": 0.00038554419885416765, + "loss": 1.8553, + "step": 3645 + }, + { + "epoch": 0.3560546875, + "grad_norm": 0.2553071081638336, + "learning_rate": 0.0003854822553222004, + "loss": 1.8828, + "step": 3646 + }, + { + "epoch": 0.35615234375, + "grad_norm": 0.20786093175411224, + "learning_rate": 0.0003854203007540345, + "loss": 1.7858, + "step": 3647 + }, + { + "epoch": 0.35625, + "grad_norm": 0.2537551820278168, + "learning_rate": 0.00038535833515585883, + "loss": 1.8407, + "step": 3648 + }, + { + "epoch": 0.35634765625, + "grad_norm": 0.23400303721427917, + "learning_rate": 0.0003852963585338631, + "loss": 1.8623, + "step": 3649 + }, + { + "epoch": 0.3564453125, + "grad_norm": 0.21759647130966187, + "learning_rate": 0.0003852343708942385, + "loss": 1.8419, + "step": 3650 + }, + { + "epoch": 0.35654296875, + "grad_norm": 0.21964114904403687, + "learning_rate": 0.00038517237224317664, + "loss": 1.8366, + "step": 3651 + }, + { + "epoch": 0.356640625, + "grad_norm": 0.21748779714107513, + "learning_rate": 0.0003851103625868709, + "loss": 1.8917, + "step": 3652 + }, + { + "epoch": 0.35673828125, + "grad_norm": 0.216062992811203, + "learning_rate": 0.00038504834193151547, + "loss": 1.87, + "step": 3653 + }, + { + "epoch": 0.3568359375, + "grad_norm": 0.223576620221138, + "learning_rate": 0.0003849863102833055, + "loss": 1.8502, + "step": 3654 + }, + { + "epoch": 0.35693359375, + "grad_norm": 0.19889536499977112, + "learning_rate": 0.0003849242676484376, + "loss": 1.8545, + "step": 3655 + }, + { + "epoch": 0.35703125, + "grad_norm": 0.2517528533935547, + "learning_rate": 0.0003848622140331092, + "loss": 1.8575, + "step": 3656 + }, + { + "epoch": 0.35712890625, + "grad_norm": 0.22014349699020386, + "learning_rate": 0.000384800149443519, + "loss": 1.9084, + "step": 3657 + }, + { + "epoch": 0.3572265625, + "grad_norm": 0.2128230482339859, + "learning_rate": 0.0003847380738858665, + "loss": 1.8927, + "step": 3658 + }, + { + "epoch": 0.35732421875, + "grad_norm": 0.2507193088531494, + "learning_rate": 0.0003846759873663526, + "loss": 1.8391, + "step": 3659 + }, + { + "epoch": 0.357421875, + "grad_norm": 0.23501361906528473, + "learning_rate": 0.00038461388989117926, + "loss": 1.8985, + "step": 3660 + }, + { + "epoch": 0.35751953125, + "grad_norm": 0.3247581422328949, + "learning_rate": 0.00038455178146654936, + "loss": 1.8718, + "step": 3661 + }, + { + "epoch": 0.3576171875, + "grad_norm": 0.22511839866638184, + "learning_rate": 0.00038448966209866703, + "loss": 1.8715, + "step": 3662 + }, + { + "epoch": 0.35771484375, + "grad_norm": 0.2150260955095291, + "learning_rate": 0.00038442753179373753, + "loss": 1.8549, + "step": 3663 + }, + { + "epoch": 0.3578125, + "grad_norm": 0.2238318771123886, + "learning_rate": 0.00038436539055796705, + "loss": 1.873, + "step": 3664 + }, + { + "epoch": 0.35791015625, + "grad_norm": 0.2875686287879944, + "learning_rate": 0.0003843032383975628, + "loss": 1.8757, + "step": 3665 + }, + { + "epoch": 0.3580078125, + "grad_norm": 0.25750482082366943, + "learning_rate": 0.00038424107531873346, + "loss": 1.8343, + "step": 3666 + }, + { + "epoch": 0.35810546875, + "grad_norm": 0.21657031774520874, + "learning_rate": 0.00038417890132768853, + "loss": 1.9179, + "step": 3667 + }, + { + "epoch": 0.358203125, + "grad_norm": 0.2154138833284378, + "learning_rate": 0.00038411671643063855, + "loss": 1.8464, + "step": 3668 + }, + { + "epoch": 0.35830078125, + "grad_norm": 0.19663992524147034, + "learning_rate": 0.00038405452063379524, + "loss": 1.8742, + "step": 3669 + }, + { + "epoch": 0.3583984375, + "grad_norm": 0.21699386835098267, + "learning_rate": 0.0003839923139433715, + "loss": 1.8815, + "step": 3670 + }, + { + "epoch": 0.35849609375, + "grad_norm": 0.23034091293811798, + "learning_rate": 0.00038393009636558125, + "loss": 1.8276, + "step": 3671 + }, + { + "epoch": 0.35859375, + "grad_norm": 0.203604593873024, + "learning_rate": 0.00038386786790663933, + "loss": 1.9034, + "step": 3672 + }, + { + "epoch": 0.35869140625, + "grad_norm": 0.23865197598934174, + "learning_rate": 0.00038380562857276185, + "loss": 1.9006, + "step": 3673 + }, + { + "epoch": 0.3587890625, + "grad_norm": 0.18150199949741364, + "learning_rate": 0.0003837433783701661, + "loss": 1.8433, + "step": 3674 + }, + { + "epoch": 0.35888671875, + "grad_norm": 0.2704537510871887, + "learning_rate": 0.00038368111730507015, + "loss": 1.8847, + "step": 3675 + }, + { + "epoch": 0.358984375, + "grad_norm": 0.19624443352222443, + "learning_rate": 0.00038361884538369345, + "loss": 1.849, + "step": 3676 + }, + { + "epoch": 0.35908203125, + "grad_norm": 0.26155540347099304, + "learning_rate": 0.00038355656261225633, + "loss": 1.8205, + "step": 3677 + }, + { + "epoch": 0.3591796875, + "grad_norm": 0.2297595739364624, + "learning_rate": 0.0003834942689969804, + "loss": 1.8437, + "step": 3678 + }, + { + "epoch": 0.35927734375, + "grad_norm": 0.2379789799451828, + "learning_rate": 0.00038343196454408815, + "loss": 1.8765, + "step": 3679 + }, + { + "epoch": 0.359375, + "grad_norm": 0.24337363243103027, + "learning_rate": 0.0003833696492598032, + "loss": 1.8568, + "step": 3680 + }, + { + "epoch": 0.35947265625, + "grad_norm": 0.23507575690746307, + "learning_rate": 0.0003833073231503505, + "loss": 1.8405, + "step": 3681 + }, + { + "epoch": 0.3595703125, + "grad_norm": 0.20884008705615997, + "learning_rate": 0.0003832449862219556, + "loss": 1.8292, + "step": 3682 + }, + { + "epoch": 0.35966796875, + "grad_norm": 0.24468238651752472, + "learning_rate": 0.0003831826384808456, + "loss": 1.8841, + "step": 3683 + }, + { + "epoch": 0.359765625, + "grad_norm": 0.19200499355793, + "learning_rate": 0.0003831202799332484, + "loss": 1.831, + "step": 3684 + }, + { + "epoch": 0.35986328125, + "grad_norm": 0.2582682967185974, + "learning_rate": 0.00038305791058539313, + "loss": 1.9164, + "step": 3685 + }, + { + "epoch": 0.3599609375, + "grad_norm": 0.2183380275964737, + "learning_rate": 0.00038299553044350977, + "loss": 1.8589, + "step": 3686 + }, + { + "epoch": 0.36005859375, + "grad_norm": 0.2560194134712219, + "learning_rate": 0.0003829331395138298, + "loss": 1.8866, + "step": 3687 + }, + { + "epoch": 0.36015625, + "grad_norm": 0.23860234022140503, + "learning_rate": 0.0003828707378025854, + "loss": 1.851, + "step": 3688 + }, + { + "epoch": 0.36025390625, + "grad_norm": 0.2541812062263489, + "learning_rate": 0.0003828083253160099, + "loss": 1.827, + "step": 3689 + }, + { + "epoch": 0.3603515625, + "grad_norm": 0.23112213611602783, + "learning_rate": 0.0003827459020603378, + "loss": 1.8759, + "step": 3690 + }, + { + "epoch": 0.36044921875, + "grad_norm": 0.2021452635526657, + "learning_rate": 0.0003826834680418047, + "loss": 1.8809, + "step": 3691 + }, + { + "epoch": 0.360546875, + "grad_norm": 0.21591801941394806, + "learning_rate": 0.00038262102326664705, + "loss": 1.8231, + "step": 3692 + }, + { + "epoch": 0.36064453125, + "grad_norm": 0.23959480226039886, + "learning_rate": 0.0003825585677411025, + "loss": 1.8422, + "step": 3693 + }, + { + "epoch": 0.3607421875, + "grad_norm": 0.22720707952976227, + "learning_rate": 0.0003824961014714101, + "loss": 1.8724, + "step": 3694 + }, + { + "epoch": 0.36083984375, + "grad_norm": 0.19454148411750793, + "learning_rate": 0.0003824336244638095, + "loss": 1.842, + "step": 3695 + }, + { + "epoch": 0.3609375, + "grad_norm": 0.31191563606262207, + "learning_rate": 0.00038237113672454147, + "loss": 1.8517, + "step": 3696 + }, + { + "epoch": 0.36103515625, + "grad_norm": 0.22866396605968475, + "learning_rate": 0.0003823086382598482, + "loss": 1.8568, + "step": 3697 + }, + { + "epoch": 0.3611328125, + "grad_norm": 0.265703946352005, + "learning_rate": 0.00038224612907597255, + "loss": 1.8486, + "step": 3698 + }, + { + "epoch": 0.36123046875, + "grad_norm": 0.24628691375255585, + "learning_rate": 0.00038218360917915885, + "loss": 1.8597, + "step": 3699 + }, + { + "epoch": 0.361328125, + "grad_norm": 0.23236118257045746, + "learning_rate": 0.00038212107857565203, + "loss": 1.8615, + "step": 3700 + }, + { + "epoch": 0.36142578125, + "grad_norm": 0.308764785528183, + "learning_rate": 0.00038205853727169855, + "loss": 1.9187, + "step": 3701 + }, + { + "epoch": 0.3615234375, + "grad_norm": 0.1871979832649231, + "learning_rate": 0.0003819959852735456, + "loss": 1.8889, + "step": 3702 + }, + { + "epoch": 0.36162109375, + "grad_norm": 0.2598519027233124, + "learning_rate": 0.0003819334225874416, + "loss": 1.8565, + "step": 3703 + }, + { + "epoch": 0.36171875, + "grad_norm": 0.2150181531906128, + "learning_rate": 0.0003818708492196361, + "loss": 1.8964, + "step": 3704 + }, + { + "epoch": 0.36181640625, + "grad_norm": 0.24690137803554535, + "learning_rate": 0.0003818082651763795, + "loss": 1.8602, + "step": 3705 + }, + { + "epoch": 0.3619140625, + "grad_norm": 0.24732674658298492, + "learning_rate": 0.0003817456704639235, + "loss": 1.8561, + "step": 3706 + }, + { + "epoch": 0.36201171875, + "grad_norm": 0.18767625093460083, + "learning_rate": 0.0003816830650885206, + "loss": 1.8208, + "step": 3707 + }, + { + "epoch": 0.362109375, + "grad_norm": 0.20773689448833466, + "learning_rate": 0.0003816204490564247, + "loss": 1.8739, + "step": 3708 + }, + { + "epoch": 0.36220703125, + "grad_norm": 0.2196119874715805, + "learning_rate": 0.00038155782237389054, + "loss": 1.8929, + "step": 3709 + }, + { + "epoch": 0.3623046875, + "grad_norm": 0.21558576822280884, + "learning_rate": 0.00038149518504717385, + "loss": 1.8732, + "step": 3710 + }, + { + "epoch": 0.36240234375, + "grad_norm": 0.20954236388206482, + "learning_rate": 0.00038143253708253173, + "loss": 1.8729, + "step": 3711 + }, + { + "epoch": 0.3625, + "grad_norm": 0.2093965858221054, + "learning_rate": 0.00038136987848622197, + "loss": 1.8957, + "step": 3712 + }, + { + "epoch": 0.36259765625, + "grad_norm": 0.21315394341945648, + "learning_rate": 0.00038130720926450384, + "loss": 1.8769, + "step": 3713 + }, + { + "epoch": 0.3626953125, + "grad_norm": 0.20709800720214844, + "learning_rate": 0.00038124452942363713, + "loss": 1.8633, + "step": 3714 + }, + { + "epoch": 0.36279296875, + "grad_norm": 0.18485292792320251, + "learning_rate": 0.0003811818389698833, + "loss": 1.8707, + "step": 3715 + }, + { + "epoch": 0.362890625, + "grad_norm": 0.21978256106376648, + "learning_rate": 0.00038111913790950437, + "loss": 1.8756, + "step": 3716 + }, + { + "epoch": 0.36298828125, + "grad_norm": 0.22362416982650757, + "learning_rate": 0.00038105642624876367, + "loss": 1.8429, + "step": 3717 + }, + { + "epoch": 0.3630859375, + "grad_norm": 0.20255053043365479, + "learning_rate": 0.0003809937039939257, + "loss": 1.8815, + "step": 3718 + }, + { + "epoch": 0.36318359375, + "grad_norm": 0.25443053245544434, + "learning_rate": 0.0003809309711512556, + "loss": 1.8544, + "step": 3719 + }, + { + "epoch": 0.36328125, + "grad_norm": 0.20008181035518646, + "learning_rate": 0.00038086822772702, + "loss": 1.8782, + "step": 3720 + }, + { + "epoch": 0.36337890625, + "grad_norm": 0.25505152344703674, + "learning_rate": 0.0003808054737274863, + "loss": 1.8917, + "step": 3721 + }, + { + "epoch": 0.3634765625, + "grad_norm": 0.23726782202720642, + "learning_rate": 0.00038074270915892315, + "loss": 1.8482, + "step": 3722 + }, + { + "epoch": 0.36357421875, + "grad_norm": 0.21997596323490143, + "learning_rate": 0.0003806799340276002, + "loss": 1.878, + "step": 3723 + }, + { + "epoch": 0.363671875, + "grad_norm": 0.27170437574386597, + "learning_rate": 0.0003806171483397881, + "loss": 1.825, + "step": 3724 + }, + { + "epoch": 0.36376953125, + "grad_norm": 0.18107932806015015, + "learning_rate": 0.00038055435210175846, + "loss": 1.8275, + "step": 3725 + }, + { + "epoch": 0.3638671875, + "grad_norm": 0.22321036458015442, + "learning_rate": 0.00038049154531978423, + "loss": 1.8339, + "step": 3726 + }, + { + "epoch": 0.36396484375, + "grad_norm": 0.24020689725875854, + "learning_rate": 0.0003804287280001392, + "loss": 1.8945, + "step": 3727 + }, + { + "epoch": 0.3640625, + "grad_norm": 0.21679888665676117, + "learning_rate": 0.0003803659001490982, + "loss": 1.871, + "step": 3728 + }, + { + "epoch": 0.36416015625, + "grad_norm": 0.20395179092884064, + "learning_rate": 0.00038030306177293726, + "loss": 1.8462, + "step": 3729 + }, + { + "epoch": 0.3642578125, + "grad_norm": 0.19003739953041077, + "learning_rate": 0.0003802402128779334, + "loss": 1.9009, + "step": 3730 + }, + { + "epoch": 0.36435546875, + "grad_norm": 0.21998478472232819, + "learning_rate": 0.00038017735347036457, + "loss": 1.8584, + "step": 3731 + }, + { + "epoch": 0.364453125, + "grad_norm": 0.19138023257255554, + "learning_rate": 0.00038011448355650996, + "loss": 1.8894, + "step": 3732 + }, + { + "epoch": 0.36455078125, + "grad_norm": 0.20978252589702606, + "learning_rate": 0.00038005160314264966, + "loss": 1.808, + "step": 3733 + }, + { + "epoch": 0.3646484375, + "grad_norm": 0.2341012954711914, + "learning_rate": 0.00037998871223506487, + "loss": 1.8356, + "step": 3734 + }, + { + "epoch": 0.36474609375, + "grad_norm": 0.21801787614822388, + "learning_rate": 0.0003799258108400378, + "loss": 1.819, + "step": 3735 + }, + { + "epoch": 0.36484375, + "grad_norm": 0.21013173460960388, + "learning_rate": 0.00037986289896385183, + "loss": 1.8275, + "step": 3736 + }, + { + "epoch": 0.36494140625, + "grad_norm": 0.24805255234241486, + "learning_rate": 0.00037979997661279123, + "loss": 1.9283, + "step": 3737 + }, + { + "epoch": 0.3650390625, + "grad_norm": 0.21214769780635834, + "learning_rate": 0.0003797370437931414, + "loss": 1.8444, + "step": 3738 + }, + { + "epoch": 0.36513671875, + "grad_norm": 0.2353964000940323, + "learning_rate": 0.0003796741005111889, + "loss": 1.8862, + "step": 3739 + }, + { + "epoch": 0.365234375, + "grad_norm": 0.17549175024032593, + "learning_rate": 0.00037961114677322093, + "loss": 1.8671, + "step": 3740 + }, + { + "epoch": 0.36533203125, + "grad_norm": 0.24111035466194153, + "learning_rate": 0.0003795481825855263, + "loss": 1.8462, + "step": 3741 + }, + { + "epoch": 0.3654296875, + "grad_norm": 0.26238951086997986, + "learning_rate": 0.00037948520795439436, + "loss": 1.892, + "step": 3742 + }, + { + "epoch": 0.36552734375, + "grad_norm": 0.21740008890628815, + "learning_rate": 0.00037942222288611584, + "loss": 1.8798, + "step": 3743 + }, + { + "epoch": 0.365625, + "grad_norm": 0.22931286692619324, + "learning_rate": 0.0003793592273869823, + "loss": 1.8866, + "step": 3744 + }, + { + "epoch": 0.36572265625, + "grad_norm": 0.19847440719604492, + "learning_rate": 0.0003792962214632865, + "loss": 1.8912, + "step": 3745 + }, + { + "epoch": 0.3658203125, + "grad_norm": 0.186073899269104, + "learning_rate": 0.00037923320512132225, + "loss": 1.741, + "step": 3746 + }, + { + "epoch": 0.36591796875, + "grad_norm": 0.21398687362670898, + "learning_rate": 0.0003791701783673841, + "loss": 1.8779, + "step": 3747 + }, + { + "epoch": 0.366015625, + "grad_norm": 0.16714359819889069, + "learning_rate": 0.00037910714120776814, + "loss": 1.8772, + "step": 3748 + }, + { + "epoch": 0.36611328125, + "grad_norm": 0.23024225234985352, + "learning_rate": 0.0003790440936487709, + "loss": 1.8416, + "step": 3749 + }, + { + "epoch": 0.3662109375, + "grad_norm": 0.22035610675811768, + "learning_rate": 0.0003789810356966906, + "loss": 1.8962, + "step": 3750 + }, + { + "epoch": 0.36630859375, + "grad_norm": 0.19260485470294952, + "learning_rate": 0.0003789179673578259, + "loss": 1.8683, + "step": 3751 + }, + { + "epoch": 0.36640625, + "grad_norm": 0.2358340322971344, + "learning_rate": 0.00037885488863847696, + "loss": 1.871, + "step": 3752 + }, + { + "epoch": 0.36650390625, + "grad_norm": 0.2558320462703705, + "learning_rate": 0.00037879179954494475, + "loss": 1.8232, + "step": 3753 + }, + { + "epoch": 0.3666015625, + "grad_norm": 0.20952382683753967, + "learning_rate": 0.00037872870008353124, + "loss": 1.8833, + "step": 3754 + }, + { + "epoch": 0.36669921875, + "grad_norm": 0.20016750693321228, + "learning_rate": 0.0003786655902605396, + "loss": 1.8461, + "step": 3755 + }, + { + "epoch": 0.366796875, + "grad_norm": 0.21693620085716248, + "learning_rate": 0.0003786024700822738, + "loss": 1.8463, + "step": 3756 + }, + { + "epoch": 0.36689453125, + "grad_norm": 0.18927980959415436, + "learning_rate": 0.0003785393395550391, + "loss": 1.8573, + "step": 3757 + }, + { + "epoch": 0.3669921875, + "grad_norm": 0.21964512765407562, + "learning_rate": 0.0003784761986851416, + "loss": 1.8861, + "step": 3758 + }, + { + "epoch": 0.36708984375, + "grad_norm": 0.2026022970676422, + "learning_rate": 0.0003784130474788887, + "loss": 1.8446, + "step": 3759 + }, + { + "epoch": 0.3671875, + "grad_norm": 0.19343462586402893, + "learning_rate": 0.00037834988594258843, + "loss": 1.8459, + "step": 3760 + }, + { + "epoch": 0.36728515625, + "grad_norm": 0.22341008484363556, + "learning_rate": 0.0003782867140825502, + "loss": 1.9195, + "step": 3761 + }, + { + "epoch": 0.3673828125, + "grad_norm": 0.21624907851219177, + "learning_rate": 0.0003782235319050842, + "loss": 1.8169, + "step": 3762 + }, + { + "epoch": 0.36748046875, + "grad_norm": 0.2739900052547455, + "learning_rate": 0.00037816033941650185, + "loss": 1.8959, + "step": 3763 + }, + { + "epoch": 0.367578125, + "grad_norm": 0.20762409269809723, + "learning_rate": 0.0003780971366231156, + "loss": 1.8501, + "step": 3764 + }, + { + "epoch": 0.36767578125, + "grad_norm": 0.23242148756980896, + "learning_rate": 0.0003780339235312387, + "loss": 1.8915, + "step": 3765 + }, + { + "epoch": 0.3677734375, + "grad_norm": 0.2940577268600464, + "learning_rate": 0.00037797070014718555, + "loss": 1.8851, + "step": 3766 + }, + { + "epoch": 0.36787109375, + "grad_norm": 0.26492175459861755, + "learning_rate": 0.0003779074664772718, + "loss": 1.8035, + "step": 3767 + }, + { + "epoch": 0.36796875, + "grad_norm": 0.23707738518714905, + "learning_rate": 0.0003778442225278138, + "loss": 1.8489, + "step": 3768 + }, + { + "epoch": 0.36806640625, + "grad_norm": 0.1993209421634674, + "learning_rate": 0.0003777809683051291, + "loss": 1.8721, + "step": 3769 + }, + { + "epoch": 0.3681640625, + "grad_norm": 0.24402756989002228, + "learning_rate": 0.0003777177038155362, + "loss": 1.8695, + "step": 3770 + }, + { + "epoch": 0.36826171875, + "grad_norm": 0.21460117399692535, + "learning_rate": 0.00037765442906535475, + "loss": 1.8673, + "step": 3771 + }, + { + "epoch": 0.368359375, + "grad_norm": 0.205267071723938, + "learning_rate": 0.0003775911440609052, + "loss": 1.8147, + "step": 3772 + }, + { + "epoch": 0.36845703125, + "grad_norm": 0.20510420203208923, + "learning_rate": 0.0003775278488085093, + "loss": 1.854, + "step": 3773 + }, + { + "epoch": 0.3685546875, + "grad_norm": 0.20449259877204895, + "learning_rate": 0.0003774645433144896, + "loss": 1.898, + "step": 3774 + }, + { + "epoch": 0.36865234375, + "grad_norm": 0.24301192164421082, + "learning_rate": 0.0003774012275851697, + "loss": 1.8672, + "step": 3775 + }, + { + "epoch": 0.36875, + "grad_norm": 0.18098856508731842, + "learning_rate": 0.00037733790162687445, + "loss": 1.8821, + "step": 3776 + }, + { + "epoch": 0.36884765625, + "grad_norm": 0.2757241427898407, + "learning_rate": 0.00037727456544592937, + "loss": 1.8566, + "step": 3777 + }, + { + "epoch": 0.3689453125, + "grad_norm": 0.20406556129455566, + "learning_rate": 0.00037721121904866133, + "loss": 1.8682, + "step": 3778 + }, + { + "epoch": 0.36904296875, + "grad_norm": 0.26397988200187683, + "learning_rate": 0.000377147862441398, + "loss": 1.8397, + "step": 3779 + }, + { + "epoch": 0.369140625, + "grad_norm": 0.22704710066318512, + "learning_rate": 0.0003770844956304682, + "loss": 1.8548, + "step": 3780 + }, + { + "epoch": 0.36923828125, + "grad_norm": 0.2295246124267578, + "learning_rate": 0.0003770211186222017, + "loss": 1.8748, + "step": 3781 + }, + { + "epoch": 0.3693359375, + "grad_norm": 0.2571696937084198, + "learning_rate": 0.0003769577314229292, + "loss": 1.8545, + "step": 3782 + }, + { + "epoch": 0.36943359375, + "grad_norm": 0.22579611837863922, + "learning_rate": 0.0003768943340389826, + "loss": 1.8948, + "step": 3783 + }, + { + "epoch": 0.36953125, + "grad_norm": 0.21409833431243896, + "learning_rate": 0.0003768309264766947, + "loss": 1.7853, + "step": 3784 + }, + { + "epoch": 0.36962890625, + "grad_norm": 0.23976057767868042, + "learning_rate": 0.0003767675087423995, + "loss": 1.8401, + "step": 3785 + }, + { + "epoch": 0.3697265625, + "grad_norm": 0.21009457111358643, + "learning_rate": 0.0003767040808424317, + "loss": 1.8773, + "step": 3786 + }, + { + "epoch": 0.36982421875, + "grad_norm": 0.27418291568756104, + "learning_rate": 0.0003766406427831272, + "loss": 1.8874, + "step": 3787 + }, + { + "epoch": 0.369921875, + "grad_norm": 0.20294521749019623, + "learning_rate": 0.000376577194570823, + "loss": 1.8804, + "step": 3788 + }, + { + "epoch": 0.37001953125, + "grad_norm": 0.3269689679145813, + "learning_rate": 0.00037651373621185697, + "loss": 1.8547, + "step": 3789 + }, + { + "epoch": 0.3701171875, + "grad_norm": 0.2686477303504944, + "learning_rate": 0.0003764502677125679, + "loss": 1.8666, + "step": 3790 + }, + { + "epoch": 0.37021484375, + "grad_norm": 0.2950127422809601, + "learning_rate": 0.000376386789079296, + "loss": 1.9096, + "step": 3791 + }, + { + "epoch": 0.3703125, + "grad_norm": 0.19845204055309296, + "learning_rate": 0.00037632330031838195, + "loss": 1.8558, + "step": 3792 + }, + { + "epoch": 0.37041015625, + "grad_norm": 0.27506914734840393, + "learning_rate": 0.00037625980143616796, + "loss": 1.8602, + "step": 3793 + }, + { + "epoch": 0.3705078125, + "grad_norm": 0.23172548413276672, + "learning_rate": 0.0003761962924389968, + "loss": 1.8983, + "step": 3794 + }, + { + "epoch": 0.37060546875, + "grad_norm": 0.19600559771060944, + "learning_rate": 0.0003761327733332126, + "loss": 1.8958, + "step": 3795 + }, + { + "epoch": 0.370703125, + "grad_norm": 0.22822436690330505, + "learning_rate": 0.0003760692441251602, + "loss": 1.8606, + "step": 3796 + }, + { + "epoch": 0.37080078125, + "grad_norm": 0.2549194097518921, + "learning_rate": 0.00037600570482118585, + "loss": 1.8825, + "step": 3797 + }, + { + "epoch": 0.3708984375, + "grad_norm": 0.21062737703323364, + "learning_rate": 0.00037594215542763624, + "loss": 1.8963, + "step": 3798 + }, + { + "epoch": 0.37099609375, + "grad_norm": 0.2261582314968109, + "learning_rate": 0.0003758785959508598, + "loss": 1.8917, + "step": 3799 + }, + { + "epoch": 0.37109375, + "grad_norm": 0.22571882605552673, + "learning_rate": 0.00037581502639720516, + "loss": 1.802, + "step": 3800 + }, + { + "epoch": 0.37119140625, + "grad_norm": 0.23393283784389496, + "learning_rate": 0.0003757514467730225, + "loss": 1.855, + "step": 3801 + }, + { + "epoch": 0.3712890625, + "grad_norm": 0.21826542913913727, + "learning_rate": 0.00037568785708466304, + "loss": 1.8376, + "step": 3802 + }, + { + "epoch": 0.37138671875, + "grad_norm": 0.19500795006752014, + "learning_rate": 0.00037562425733847856, + "loss": 1.8535, + "step": 3803 + }, + { + "epoch": 0.371484375, + "grad_norm": 0.22810782492160797, + "learning_rate": 0.00037556064754082233, + "loss": 1.8501, + "step": 3804 + }, + { + "epoch": 0.37158203125, + "grad_norm": 0.17076687514781952, + "learning_rate": 0.00037549702769804815, + "loss": 1.8216, + "step": 3805 + }, + { + "epoch": 0.3716796875, + "grad_norm": 0.21661575138568878, + "learning_rate": 0.00037543339781651134, + "loss": 1.8498, + "step": 3806 + }, + { + "epoch": 0.37177734375, + "grad_norm": 0.24469956755638123, + "learning_rate": 0.00037536975790256776, + "loss": 1.8986, + "step": 3807 + }, + { + "epoch": 0.371875, + "grad_norm": 0.2178879976272583, + "learning_rate": 0.0003753061079625746, + "loss": 1.8269, + "step": 3808 + }, + { + "epoch": 0.37197265625, + "grad_norm": 0.19541752338409424, + "learning_rate": 0.00037524244800288985, + "loss": 1.8767, + "step": 3809 + }, + { + "epoch": 0.3720703125, + "grad_norm": 0.22389501333236694, + "learning_rate": 0.0003751787780298727, + "loss": 1.8638, + "step": 3810 + }, + { + "epoch": 0.37216796875, + "grad_norm": 0.2582794427871704, + "learning_rate": 0.0003751150980498831, + "loss": 1.8045, + "step": 3811 + }, + { + "epoch": 0.372265625, + "grad_norm": 0.22942645847797394, + "learning_rate": 0.00037505140806928214, + "loss": 1.8611, + "step": 3812 + }, + { + "epoch": 0.37236328125, + "grad_norm": 0.20219996571540833, + "learning_rate": 0.00037498770809443185, + "loss": 1.8577, + "step": 3813 + }, + { + "epoch": 0.3724609375, + "grad_norm": 0.21832026541233063, + "learning_rate": 0.00037492399813169534, + "loss": 1.8651, + "step": 3814 + }, + { + "epoch": 0.37255859375, + "grad_norm": 0.2228987216949463, + "learning_rate": 0.00037486027818743665, + "loss": 1.8622, + "step": 3815 + }, + { + "epoch": 0.37265625, + "grad_norm": 0.24790452420711517, + "learning_rate": 0.0003747965482680209, + "loss": 1.8527, + "step": 3816 + }, + { + "epoch": 0.37275390625, + "grad_norm": 0.19790831208229065, + "learning_rate": 0.0003747328083798141, + "loss": 1.9084, + "step": 3817 + }, + { + "epoch": 0.3728515625, + "grad_norm": 0.21211573481559753, + "learning_rate": 0.00037466905852918324, + "loss": 1.8931, + "step": 3818 + }, + { + "epoch": 0.37294921875, + "grad_norm": 0.20081430673599243, + "learning_rate": 0.0003746052987224964, + "loss": 1.7665, + "step": 3819 + }, + { + "epoch": 0.373046875, + "grad_norm": 0.19182641804218292, + "learning_rate": 0.0003745415289661228, + "loss": 1.8826, + "step": 3820 + }, + { + "epoch": 0.37314453125, + "grad_norm": 0.1944439709186554, + "learning_rate": 0.0003744777492664321, + "loss": 1.8696, + "step": 3821 + }, + { + "epoch": 0.3732421875, + "grad_norm": 0.17847688496112823, + "learning_rate": 0.00037441395962979575, + "loss": 1.8371, + "step": 3822 + }, + { + "epoch": 0.37333984375, + "grad_norm": 0.19860583543777466, + "learning_rate": 0.00037435016006258544, + "loss": 1.8642, + "step": 3823 + }, + { + "epoch": 0.3734375, + "grad_norm": 0.18326810002326965, + "learning_rate": 0.0003742863505711744, + "loss": 1.8481, + "step": 3824 + }, + { + "epoch": 0.37353515625, + "grad_norm": 0.2152925431728363, + "learning_rate": 0.0003742225311619364, + "loss": 1.8934, + "step": 3825 + }, + { + "epoch": 0.3736328125, + "grad_norm": 0.20573176443576813, + "learning_rate": 0.00037415870184124666, + "loss": 1.8624, + "step": 3826 + }, + { + "epoch": 0.37373046875, + "grad_norm": 0.23908239603042603, + "learning_rate": 0.00037409486261548105, + "loss": 1.8815, + "step": 3827 + }, + { + "epoch": 0.373828125, + "grad_norm": 0.21868613362312317, + "learning_rate": 0.0003740310134910165, + "loss": 1.8917, + "step": 3828 + }, + { + "epoch": 0.37392578125, + "grad_norm": 0.22748807072639465, + "learning_rate": 0.00037396715447423116, + "loss": 1.8948, + "step": 3829 + }, + { + "epoch": 0.3740234375, + "grad_norm": 0.2231343686580658, + "learning_rate": 0.00037390328557150376, + "loss": 1.8595, + "step": 3830 + }, + { + "epoch": 0.37412109375, + "grad_norm": 0.19311115145683289, + "learning_rate": 0.0003738394067892144, + "loss": 1.833, + "step": 3831 + }, + { + "epoch": 0.37421875, + "grad_norm": 0.24683715403079987, + "learning_rate": 0.00037377551813374394, + "loss": 1.8881, + "step": 3832 + }, + { + "epoch": 0.37431640625, + "grad_norm": 0.27622419595718384, + "learning_rate": 0.0003737116196114742, + "loss": 1.8784, + "step": 3833 + }, + { + "epoch": 0.3744140625, + "grad_norm": 0.20628823339939117, + "learning_rate": 0.0003736477112287883, + "loss": 1.856, + "step": 3834 + }, + { + "epoch": 0.37451171875, + "grad_norm": 0.2787981927394867, + "learning_rate": 0.0003735837929920698, + "loss": 1.8345, + "step": 3835 + }, + { + "epoch": 0.374609375, + "grad_norm": 0.25385451316833496, + "learning_rate": 0.00037351986490770394, + "loss": 1.8824, + "step": 3836 + }, + { + "epoch": 0.37470703125, + "grad_norm": 0.2456686645746231, + "learning_rate": 0.0003734559269820763, + "loss": 1.8494, + "step": 3837 + }, + { + "epoch": 0.3748046875, + "grad_norm": 0.23777589201927185, + "learning_rate": 0.0003733919792215738, + "loss": 1.8332, + "step": 3838 + }, + { + "epoch": 0.37490234375, + "grad_norm": 0.24769902229309082, + "learning_rate": 0.0003733280216325843, + "loss": 1.8698, + "step": 3839 + }, + { + "epoch": 0.375, + "grad_norm": 0.23901763558387756, + "learning_rate": 0.0003732640542214965, + "loss": 1.8203, + "step": 3840 + }, + { + "epoch": 0.37509765625, + "grad_norm": 0.20374193787574768, + "learning_rate": 0.0003732000769947003, + "loss": 1.8252, + "step": 3841 + }, + { + "epoch": 0.3751953125, + "grad_norm": 0.2926974594593048, + "learning_rate": 0.00037313608995858625, + "loss": 1.8683, + "step": 3842 + }, + { + "epoch": 0.37529296875, + "grad_norm": 0.19516809284687042, + "learning_rate": 0.00037307209311954626, + "loss": 1.8714, + "step": 3843 + }, + { + "epoch": 0.375390625, + "grad_norm": 0.2639426589012146, + "learning_rate": 0.00037300808648397306, + "loss": 1.8392, + "step": 3844 + }, + { + "epoch": 0.37548828125, + "grad_norm": 0.2614942193031311, + "learning_rate": 0.00037294407005826024, + "loss": 1.8272, + "step": 3845 + }, + { + "epoch": 0.3755859375, + "grad_norm": 0.1904597282409668, + "learning_rate": 0.0003728800438488026, + "loss": 1.862, + "step": 3846 + }, + { + "epoch": 0.37568359375, + "grad_norm": 0.2732922434806824, + "learning_rate": 0.00037281600786199566, + "loss": 1.86, + "step": 3847 + }, + { + "epoch": 0.37578125, + "grad_norm": 0.19078542292118073, + "learning_rate": 0.0003727519621042361, + "loss": 1.8619, + "step": 3848 + }, + { + "epoch": 0.37587890625, + "grad_norm": 0.2331223487854004, + "learning_rate": 0.00037268790658192153, + "loss": 1.8451, + "step": 3849 + }, + { + "epoch": 0.3759765625, + "grad_norm": 0.2795408070087433, + "learning_rate": 0.00037262384130145053, + "loss": 1.8958, + "step": 3850 + }, + { + "epoch": 0.37607421875, + "grad_norm": 0.18555857241153717, + "learning_rate": 0.0003725597662692227, + "loss": 1.8379, + "step": 3851 + }, + { + "epoch": 0.376171875, + "grad_norm": 0.23913845419883728, + "learning_rate": 0.00037249568149163854, + "loss": 1.7977, + "step": 3852 + }, + { + "epoch": 0.37626953125, + "grad_norm": 0.25755074620246887, + "learning_rate": 0.0003724315869750995, + "loss": 1.8633, + "step": 3853 + }, + { + "epoch": 0.3763671875, + "grad_norm": 0.23956407606601715, + "learning_rate": 0.00037236748272600806, + "loss": 1.8842, + "step": 3854 + }, + { + "epoch": 0.37646484375, + "grad_norm": 0.21027059853076935, + "learning_rate": 0.00037230336875076776, + "loss": 1.8684, + "step": 3855 + }, + { + "epoch": 0.3765625, + "grad_norm": 0.20290538668632507, + "learning_rate": 0.0003722392450557828, + "loss": 1.8714, + "step": 3856 + }, + { + "epoch": 0.37666015625, + "grad_norm": 0.24776442348957062, + "learning_rate": 0.0003721751116474589, + "loss": 1.8192, + "step": 3857 + }, + { + "epoch": 0.3767578125, + "grad_norm": 0.28582367300987244, + "learning_rate": 0.0003721109685322022, + "loss": 1.8696, + "step": 3858 + }, + { + "epoch": 0.37685546875, + "grad_norm": 0.2176811397075653, + "learning_rate": 0.00037204681571642004, + "loss": 1.8502, + "step": 3859 + }, + { + "epoch": 0.376953125, + "grad_norm": 0.2929898500442505, + "learning_rate": 0.00037198265320652064, + "loss": 1.8552, + "step": 3860 + }, + { + "epoch": 0.37705078125, + "grad_norm": 0.2156658172607422, + "learning_rate": 0.0003719184810089135, + "loss": 1.8439, + "step": 3861 + }, + { + "epoch": 0.3771484375, + "grad_norm": 0.272320955991745, + "learning_rate": 0.00037185429913000865, + "loss": 1.8302, + "step": 3862 + }, + { + "epoch": 0.37724609375, + "grad_norm": 0.21699823439121246, + "learning_rate": 0.00037179010757621745, + "loss": 1.8717, + "step": 3863 + }, + { + "epoch": 0.37734375, + "grad_norm": 0.2614113390445709, + "learning_rate": 0.00037172590635395196, + "loss": 1.8684, + "step": 3864 + }, + { + "epoch": 0.37744140625, + "grad_norm": 0.21246129274368286, + "learning_rate": 0.0003716616954696254, + "loss": 1.8775, + "step": 3865 + }, + { + "epoch": 0.3775390625, + "grad_norm": 0.24341346323490143, + "learning_rate": 0.0003715974749296517, + "loss": 1.8622, + "step": 3866 + }, + { + "epoch": 0.37763671875, + "grad_norm": 0.20468126237392426, + "learning_rate": 0.00037153324474044607, + "loss": 1.8373, + "step": 3867 + }, + { + "epoch": 0.377734375, + "grad_norm": 0.22117452323436737, + "learning_rate": 0.00037146900490842445, + "loss": 1.8322, + "step": 3868 + }, + { + "epoch": 0.37783203125, + "grad_norm": 0.21013975143432617, + "learning_rate": 0.00037140475544000394, + "loss": 1.8814, + "step": 3869 + }, + { + "epoch": 0.3779296875, + "grad_norm": 0.19857852160930634, + "learning_rate": 0.0003713404963416024, + "loss": 1.8798, + "step": 3870 + }, + { + "epoch": 0.37802734375, + "grad_norm": 0.2867245078086853, + "learning_rate": 0.0003712762276196389, + "loss": 1.8232, + "step": 3871 + }, + { + "epoch": 0.378125, + "grad_norm": 0.260097861289978, + "learning_rate": 0.000371211949280533, + "loss": 1.843, + "step": 3872 + }, + { + "epoch": 0.37822265625, + "grad_norm": 0.23447772860527039, + "learning_rate": 0.00037114766133070586, + "loss": 1.8219, + "step": 3873 + }, + { + "epoch": 0.3783203125, + "grad_norm": 0.223851278424263, + "learning_rate": 0.0003710833637765791, + "loss": 1.8897, + "step": 3874 + }, + { + "epoch": 0.37841796875, + "grad_norm": 0.2399793565273285, + "learning_rate": 0.00037101905662457546, + "loss": 1.8636, + "step": 3875 + }, + { + "epoch": 0.378515625, + "grad_norm": 0.20454320311546326, + "learning_rate": 0.00037095473988111883, + "loss": 1.8749, + "step": 3876 + }, + { + "epoch": 0.37861328125, + "grad_norm": 0.22218739986419678, + "learning_rate": 0.0003708904135526337, + "loss": 1.8741, + "step": 3877 + }, + { + "epoch": 0.3787109375, + "grad_norm": 0.2315162718296051, + "learning_rate": 0.00037082607764554574, + "loss": 1.8579, + "step": 3878 + }, + { + "epoch": 0.37880859375, + "grad_norm": 0.21069717407226562, + "learning_rate": 0.00037076173216628165, + "loss": 1.9328, + "step": 3879 + }, + { + "epoch": 0.37890625, + "grad_norm": 0.23605448007583618, + "learning_rate": 0.00037069737712126886, + "loss": 1.8225, + "step": 3880 + }, + { + "epoch": 0.37900390625, + "grad_norm": 0.18359407782554626, + "learning_rate": 0.00037063301251693593, + "loss": 1.8626, + "step": 3881 + }, + { + "epoch": 0.3791015625, + "grad_norm": 0.2351130247116089, + "learning_rate": 0.00037056863835971214, + "loss": 1.8855, + "step": 3882 + }, + { + "epoch": 0.37919921875, + "grad_norm": 0.1951916217803955, + "learning_rate": 0.0003705042546560282, + "loss": 1.8423, + "step": 3883 + }, + { + "epoch": 0.379296875, + "grad_norm": 0.2034117877483368, + "learning_rate": 0.00037043986141231525, + "loss": 1.8542, + "step": 3884 + }, + { + "epoch": 0.37939453125, + "grad_norm": 0.2252059429883957, + "learning_rate": 0.00037037545863500567, + "loss": 1.8623, + "step": 3885 + }, + { + "epoch": 0.3794921875, + "grad_norm": 0.17208275198936462, + "learning_rate": 0.0003703110463305327, + "loss": 1.8129, + "step": 3886 + }, + { + "epoch": 0.37958984375, + "grad_norm": 0.19508133828639984, + "learning_rate": 0.0003702466245053306, + "loss": 1.8187, + "step": 3887 + }, + { + "epoch": 0.3796875, + "grad_norm": 0.18515267968177795, + "learning_rate": 0.0003701821931658345, + "loss": 1.8416, + "step": 3888 + }, + { + "epoch": 0.37978515625, + "grad_norm": 0.16859371960163116, + "learning_rate": 0.0003701177523184806, + "loss": 1.9132, + "step": 3889 + }, + { + "epoch": 0.3798828125, + "grad_norm": 0.1977667212486267, + "learning_rate": 0.0003700533019697059, + "loss": 1.8912, + "step": 3890 + }, + { + "epoch": 0.37998046875, + "grad_norm": 0.18891897797584534, + "learning_rate": 0.0003699888421259483, + "loss": 1.8595, + "step": 3891 + }, + { + "epoch": 0.380078125, + "grad_norm": 0.21774737536907196, + "learning_rate": 0.00036992437279364707, + "loss": 1.911, + "step": 3892 + }, + { + "epoch": 0.38017578125, + "grad_norm": 0.29687103629112244, + "learning_rate": 0.00036985989397924187, + "loss": 1.8465, + "step": 3893 + }, + { + "epoch": 0.3802734375, + "grad_norm": 0.30139657855033875, + "learning_rate": 0.0003697954056891736, + "loss": 1.8941, + "step": 3894 + }, + { + "epoch": 0.38037109375, + "grad_norm": 0.22587136924266815, + "learning_rate": 0.00036973090792988424, + "loss": 1.8231, + "step": 3895 + }, + { + "epoch": 0.38046875, + "grad_norm": 0.2205764204263687, + "learning_rate": 0.0003696664007078163, + "loss": 1.8506, + "step": 3896 + }, + { + "epoch": 0.38056640625, + "grad_norm": 0.1905178278684616, + "learning_rate": 0.00036960188402941375, + "loss": 1.8633, + "step": 3897 + }, + { + "epoch": 0.3806640625, + "grad_norm": 0.24393075704574585, + "learning_rate": 0.00036953735790112087, + "loss": 1.8828, + "step": 3898 + }, + { + "epoch": 0.38076171875, + "grad_norm": 0.23119643330574036, + "learning_rate": 0.0003694728223293836, + "loss": 1.8468, + "step": 3899 + }, + { + "epoch": 0.380859375, + "grad_norm": 0.274705708026886, + "learning_rate": 0.0003694082773206483, + "loss": 1.889, + "step": 3900 + }, + { + "epoch": 0.38095703125, + "grad_norm": 0.18551072478294373, + "learning_rate": 0.0003693437228813624, + "loss": 1.8425, + "step": 3901 + }, + { + "epoch": 0.3810546875, + "grad_norm": 0.3081590533256531, + "learning_rate": 0.0003692791590179746, + "loss": 1.8491, + "step": 3902 + }, + { + "epoch": 0.38115234375, + "grad_norm": 0.20071613788604736, + "learning_rate": 0.00036921458573693387, + "loss": 1.8793, + "step": 3903 + }, + { + "epoch": 0.38125, + "grad_norm": 0.26677054166793823, + "learning_rate": 0.0003691500030446908, + "loss": 1.9019, + "step": 3904 + }, + { + "epoch": 0.38134765625, + "grad_norm": 0.19375574588775635, + "learning_rate": 0.0003690854109476964, + "loss": 1.8623, + "step": 3905 + }, + { + "epoch": 0.3814453125, + "grad_norm": 0.23315277695655823, + "learning_rate": 0.00036902080945240307, + "loss": 1.8494, + "step": 3906 + }, + { + "epoch": 0.38154296875, + "grad_norm": 0.24586628377437592, + "learning_rate": 0.00036895619856526377, + "loss": 1.8783, + "step": 3907 + }, + { + "epoch": 0.381640625, + "grad_norm": 0.18601994216442108, + "learning_rate": 0.00036889157829273265, + "loss": 1.7878, + "step": 3908 + }, + { + "epoch": 0.38173828125, + "grad_norm": 0.2880648970603943, + "learning_rate": 0.0003688269486412646, + "loss": 1.8621, + "step": 3909 + }, + { + "epoch": 0.3818359375, + "grad_norm": 0.20966652035713196, + "learning_rate": 0.00036876230961731564, + "loss": 1.8271, + "step": 3910 + }, + { + "epoch": 0.38193359375, + "grad_norm": 0.22924384474754333, + "learning_rate": 0.0003686976612273427, + "loss": 1.8308, + "step": 3911 + }, + { + "epoch": 0.38203125, + "grad_norm": 0.2276628315448761, + "learning_rate": 0.0003686330034778034, + "loss": 1.8565, + "step": 3912 + }, + { + "epoch": 0.38212890625, + "grad_norm": 0.2096962332725525, + "learning_rate": 0.00036856833637515657, + "loss": 1.8429, + "step": 3913 + }, + { + "epoch": 0.3822265625, + "grad_norm": 0.25153499841690063, + "learning_rate": 0.0003685036599258619, + "loss": 1.8352, + "step": 3914 + }, + { + "epoch": 0.38232421875, + "grad_norm": 0.21400980651378632, + "learning_rate": 0.00036843897413637986, + "loss": 1.9016, + "step": 3915 + }, + { + "epoch": 0.382421875, + "grad_norm": 0.2173420935869217, + "learning_rate": 0.00036837427901317225, + "loss": 1.861, + "step": 3916 + }, + { + "epoch": 0.38251953125, + "grad_norm": 0.1773025095462799, + "learning_rate": 0.0003683095745627014, + "loss": 1.849, + "step": 3917 + }, + { + "epoch": 0.3826171875, + "grad_norm": 0.20093916356563568, + "learning_rate": 0.00036824486079143064, + "loss": 1.8404, + "step": 3918 + }, + { + "epoch": 0.38271484375, + "grad_norm": 0.20080716907978058, + "learning_rate": 0.00036818013770582424, + "loss": 1.8567, + "step": 3919 + }, + { + "epoch": 0.3828125, + "grad_norm": 0.24920229613780975, + "learning_rate": 0.0003681154053123478, + "loss": 1.857, + "step": 3920 + }, + { + "epoch": 0.38291015625, + "grad_norm": 0.19836358726024628, + "learning_rate": 0.00036805066361746716, + "loss": 1.8674, + "step": 3921 + }, + { + "epoch": 0.3830078125, + "grad_norm": 0.2825253903865814, + "learning_rate": 0.00036798591262764965, + "loss": 1.8862, + "step": 3922 + }, + { + "epoch": 0.38310546875, + "grad_norm": 0.25824931263923645, + "learning_rate": 0.00036792115234936316, + "loss": 1.8765, + "step": 3923 + }, + { + "epoch": 0.383203125, + "grad_norm": 0.22667564451694489, + "learning_rate": 0.00036785638278907687, + "loss": 1.8585, + "step": 3924 + }, + { + "epoch": 0.38330078125, + "grad_norm": 0.29045218229293823, + "learning_rate": 0.0003677916039532605, + "loss": 1.858, + "step": 3925 + }, + { + "epoch": 0.3833984375, + "grad_norm": 0.20461486279964447, + "learning_rate": 0.00036772681584838497, + "loss": 1.8507, + "step": 3926 + }, + { + "epoch": 0.38349609375, + "grad_norm": 0.25927817821502686, + "learning_rate": 0.000367662018480922, + "loss": 1.8297, + "step": 3927 + }, + { + "epoch": 0.38359375, + "grad_norm": 0.22330361604690552, + "learning_rate": 0.00036759721185734433, + "loss": 1.8351, + "step": 3928 + }, + { + "epoch": 0.38369140625, + "grad_norm": 0.26381710171699524, + "learning_rate": 0.00036753239598412554, + "loss": 1.9159, + "step": 3929 + }, + { + "epoch": 0.3837890625, + "grad_norm": 0.2930384576320648, + "learning_rate": 0.0003674675708677401, + "loss": 1.857, + "step": 3930 + }, + { + "epoch": 0.38388671875, + "grad_norm": 0.19331973791122437, + "learning_rate": 0.0003674027365146636, + "loss": 1.8628, + "step": 3931 + }, + { + "epoch": 0.383984375, + "grad_norm": 0.30140820145606995, + "learning_rate": 0.00036733789293137226, + "loss": 1.8753, + "step": 3932 + }, + { + "epoch": 0.38408203125, + "grad_norm": 0.306213915348053, + "learning_rate": 0.0003672730401243435, + "loss": 1.8682, + "step": 3933 + }, + { + "epoch": 0.3841796875, + "grad_norm": 0.22284962236881256, + "learning_rate": 0.00036720817810005554, + "loss": 1.8159, + "step": 3934 + }, + { + "epoch": 0.38427734375, + "grad_norm": 0.3082934319972992, + "learning_rate": 0.0003671433068649874, + "loss": 1.893, + "step": 3935 + }, + { + "epoch": 0.384375, + "grad_norm": 0.264959454536438, + "learning_rate": 0.0003670784264256192, + "loss": 1.8645, + "step": 3936 + }, + { + "epoch": 0.38447265625, + "grad_norm": 0.20448656380176544, + "learning_rate": 0.000367013536788432, + "loss": 1.8722, + "step": 3937 + }, + { + "epoch": 0.3845703125, + "grad_norm": 0.2482098639011383, + "learning_rate": 0.0003669486379599077, + "loss": 1.8509, + "step": 3938 + }, + { + "epoch": 0.38466796875, + "grad_norm": 0.24799376726150513, + "learning_rate": 0.000366883729946529, + "loss": 1.8262, + "step": 3939 + }, + { + "epoch": 0.384765625, + "grad_norm": 0.21177054941654205, + "learning_rate": 0.0003668188127547796, + "loss": 1.8538, + "step": 3940 + }, + { + "epoch": 0.38486328125, + "grad_norm": 0.2216721624135971, + "learning_rate": 0.0003667538863911444, + "loss": 1.8115, + "step": 3941 + }, + { + "epoch": 0.3849609375, + "grad_norm": 0.2219676971435547, + "learning_rate": 0.00036668895086210867, + "loss": 1.852, + "step": 3942 + }, + { + "epoch": 0.38505859375, + "grad_norm": 0.1981191635131836, + "learning_rate": 0.00036662400617415913, + "loss": 1.8699, + "step": 3943 + }, + { + "epoch": 0.38515625, + "grad_norm": 0.19222186505794525, + "learning_rate": 0.00036655905233378306, + "loss": 1.864, + "step": 3944 + }, + { + "epoch": 0.38525390625, + "grad_norm": 0.23437833786010742, + "learning_rate": 0.00036649408934746885, + "loss": 1.9019, + "step": 3945 + }, + { + "epoch": 0.3853515625, + "grad_norm": 0.18790283799171448, + "learning_rate": 0.00036642911722170563, + "loss": 1.8592, + "step": 3946 + }, + { + "epoch": 0.38544921875, + "grad_norm": 0.2296464890241623, + "learning_rate": 0.00036636413596298365, + "loss": 1.8829, + "step": 3947 + }, + { + "epoch": 0.385546875, + "grad_norm": 0.21702931821346283, + "learning_rate": 0.0003662991455777938, + "loss": 1.868, + "step": 3948 + }, + { + "epoch": 0.38564453125, + "grad_norm": 0.1856737732887268, + "learning_rate": 0.0003662341460726282, + "loss": 1.8538, + "step": 3949 + }, + { + "epoch": 0.3857421875, + "grad_norm": 0.2527240216732025, + "learning_rate": 0.0003661691374539797, + "loss": 1.8328, + "step": 3950 + }, + { + "epoch": 0.38583984375, + "grad_norm": 0.17746932804584503, + "learning_rate": 0.00036610411972834195, + "loss": 1.8069, + "step": 3951 + }, + { + "epoch": 0.3859375, + "grad_norm": 0.19036374986171722, + "learning_rate": 0.00036603909290220993, + "loss": 1.8435, + "step": 3952 + }, + { + "epoch": 0.38603515625, + "grad_norm": 0.18192900717258453, + "learning_rate": 0.0003659740569820789, + "loss": 1.7831, + "step": 3953 + }, + { + "epoch": 0.3861328125, + "grad_norm": 0.21768184006214142, + "learning_rate": 0.0003659090119744456, + "loss": 1.8316, + "step": 3954 + }, + { + "epoch": 0.38623046875, + "grad_norm": 0.2052927315235138, + "learning_rate": 0.0003658439578858074, + "loss": 1.8758, + "step": 3955 + }, + { + "epoch": 0.386328125, + "grad_norm": 0.1776941865682602, + "learning_rate": 0.0003657788947226626, + "loss": 1.8418, + "step": 3956 + }, + { + "epoch": 0.38642578125, + "grad_norm": 0.23303310573101044, + "learning_rate": 0.0003657138224915104, + "loss": 1.8663, + "step": 3957 + }, + { + "epoch": 0.3865234375, + "grad_norm": 0.20003245770931244, + "learning_rate": 0.000365648741198851, + "loss": 1.8711, + "step": 3958 + }, + { + "epoch": 0.38662109375, + "grad_norm": 0.19043077528476715, + "learning_rate": 0.00036558365085118545, + "loss": 1.8747, + "step": 3959 + }, + { + "epoch": 0.38671875, + "grad_norm": 0.18690763413906097, + "learning_rate": 0.00036551855145501566, + "loss": 1.8116, + "step": 3960 + }, + { + "epoch": 0.38681640625, + "grad_norm": 0.25892189145088196, + "learning_rate": 0.0003654534430168445, + "loss": 1.8288, + "step": 3961 + }, + { + "epoch": 0.3869140625, + "grad_norm": 0.2468033730983734, + "learning_rate": 0.0003653883255431758, + "loss": 1.8459, + "step": 3962 + }, + { + "epoch": 0.38701171875, + "grad_norm": 0.1726692169904709, + "learning_rate": 0.0003653231990405141, + "loss": 1.8456, + "step": 3963 + }, + { + "epoch": 0.387109375, + "grad_norm": 0.21686433255672455, + "learning_rate": 0.00036525806351536493, + "loss": 1.8196, + "step": 3964 + }, + { + "epoch": 0.38720703125, + "grad_norm": 0.23522233963012695, + "learning_rate": 0.00036519291897423495, + "loss": 1.8325, + "step": 3965 + }, + { + "epoch": 0.3873046875, + "grad_norm": 0.2262456715106964, + "learning_rate": 0.00036512776542363135, + "loss": 1.831, + "step": 3966 + }, + { + "epoch": 0.38740234375, + "grad_norm": 0.2011958360671997, + "learning_rate": 0.0003650626028700625, + "loss": 1.8707, + "step": 3967 + }, + { + "epoch": 0.3875, + "grad_norm": 0.297262579202652, + "learning_rate": 0.0003649974313200374, + "loss": 1.8487, + "step": 3968 + }, + { + "epoch": 0.38759765625, + "grad_norm": 0.2600073516368866, + "learning_rate": 0.00036493225078006625, + "loss": 1.8283, + "step": 3969 + }, + { + "epoch": 0.3876953125, + "grad_norm": 0.26020193099975586, + "learning_rate": 0.00036486706125666, + "loss": 1.853, + "step": 3970 + }, + { + "epoch": 0.38779296875, + "grad_norm": 0.3049705922603607, + "learning_rate": 0.0003648018627563305, + "loss": 1.8563, + "step": 3971 + }, + { + "epoch": 0.387890625, + "grad_norm": 0.22959887981414795, + "learning_rate": 0.0003647366552855905, + "loss": 1.8756, + "step": 3972 + }, + { + "epoch": 0.38798828125, + "grad_norm": 0.2265494018793106, + "learning_rate": 0.0003646714388509536, + "loss": 1.8164, + "step": 3973 + }, + { + "epoch": 0.3880859375, + "grad_norm": 0.22998474538326263, + "learning_rate": 0.0003646062134589343, + "loss": 1.8905, + "step": 3974 + }, + { + "epoch": 0.38818359375, + "grad_norm": 0.19562913477420807, + "learning_rate": 0.00036454097911604817, + "loss": 1.864, + "step": 3975 + }, + { + "epoch": 0.38828125, + "grad_norm": 0.23625966906547546, + "learning_rate": 0.0003644757358288116, + "loss": 1.8254, + "step": 3976 + }, + { + "epoch": 0.38837890625, + "grad_norm": 0.25710153579711914, + "learning_rate": 0.00036441048360374154, + "loss": 1.8971, + "step": 3977 + }, + { + "epoch": 0.3884765625, + "grad_norm": 0.22964118421077728, + "learning_rate": 0.0003643452224473563, + "loss": 1.8228, + "step": 3978 + }, + { + "epoch": 0.38857421875, + "grad_norm": 0.22172428667545319, + "learning_rate": 0.00036427995236617487, + "loss": 1.8451, + "step": 3979 + }, + { + "epoch": 0.388671875, + "grad_norm": 0.184449702501297, + "learning_rate": 0.0003642146733667172, + "loss": 1.8486, + "step": 3980 + }, + { + "epoch": 0.38876953125, + "grad_norm": 0.2299502044916153, + "learning_rate": 0.000364149385455504, + "loss": 1.8722, + "step": 3981 + }, + { + "epoch": 0.3888671875, + "grad_norm": 0.20748716592788696, + "learning_rate": 0.00036408408863905696, + "loss": 1.8757, + "step": 3982 + }, + { + "epoch": 0.38896484375, + "grad_norm": 0.20994932949543, + "learning_rate": 0.0003640187829238988, + "loss": 1.8664, + "step": 3983 + }, + { + "epoch": 0.3890625, + "grad_norm": 0.20542369782924652, + "learning_rate": 0.0003639534683165527, + "loss": 1.8165, + "step": 3984 + }, + { + "epoch": 0.38916015625, + "grad_norm": 0.22056767344474792, + "learning_rate": 0.00036388814482354333, + "loss": 1.8286, + "step": 3985 + }, + { + "epoch": 0.3892578125, + "grad_norm": 0.24840861558914185, + "learning_rate": 0.00036382281245139563, + "loss": 1.8559, + "step": 3986 + }, + { + "epoch": 0.38935546875, + "grad_norm": 0.20917242765426636, + "learning_rate": 0.0003637574712066359, + "loss": 1.8134, + "step": 3987 + }, + { + "epoch": 0.389453125, + "grad_norm": 0.21101856231689453, + "learning_rate": 0.0003636921210957912, + "loss": 1.8487, + "step": 3988 + }, + { + "epoch": 0.38955078125, + "grad_norm": 0.2077265977859497, + "learning_rate": 0.00036362676212538925, + "loss": 1.8672, + "step": 3989 + }, + { + "epoch": 0.3896484375, + "grad_norm": 0.21952399611473083, + "learning_rate": 0.00036356139430195907, + "loss": 1.8617, + "step": 3990 + }, + { + "epoch": 0.38974609375, + "grad_norm": 0.2540169060230255, + "learning_rate": 0.00036349601763203, + "loss": 1.8471, + "step": 3991 + }, + { + "epoch": 0.38984375, + "grad_norm": 0.18410301208496094, + "learning_rate": 0.00036343063212213283, + "loss": 1.8571, + "step": 3992 + }, + { + "epoch": 0.38994140625, + "grad_norm": 0.22891919314861298, + "learning_rate": 0.0003633652377787989, + "loss": 1.8226, + "step": 3993 + }, + { + "epoch": 0.3900390625, + "grad_norm": 0.2170734405517578, + "learning_rate": 0.0003632998346085607, + "loss": 1.8544, + "step": 3994 + }, + { + "epoch": 0.39013671875, + "grad_norm": 0.2172975242137909, + "learning_rate": 0.00036323442261795115, + "loss": 1.8275, + "step": 3995 + }, + { + "epoch": 0.390234375, + "grad_norm": 0.22081336379051208, + "learning_rate": 0.0003631690018135045, + "loss": 1.8478, + "step": 3996 + }, + { + "epoch": 0.39033203125, + "grad_norm": 0.17948199808597565, + "learning_rate": 0.00036310357220175574, + "loss": 1.8671, + "step": 3997 + }, + { + "epoch": 0.3904296875, + "grad_norm": 0.2276734560728073, + "learning_rate": 0.0003630381337892406, + "loss": 1.8256, + "step": 3998 + }, + { + "epoch": 0.39052734375, + "grad_norm": 0.21136391162872314, + "learning_rate": 0.0003629726865824958, + "loss": 1.8463, + "step": 3999 + }, + { + "epoch": 0.390625, + "grad_norm": 0.26860180497169495, + "learning_rate": 0.000362907230588059, + "loss": 1.8529, + "step": 4000 + }, + { + "epoch": 0.39072265625, + "grad_norm": 0.22392459213733673, + "learning_rate": 0.00036284176581246876, + "loss": 1.8282, + "step": 4001 + }, + { + "epoch": 0.3908203125, + "grad_norm": 0.22236575186252594, + "learning_rate": 0.00036277629226226423, + "loss": 1.8288, + "step": 4002 + }, + { + "epoch": 0.39091796875, + "grad_norm": 0.2793997526168823, + "learning_rate": 0.0003627108099439857, + "loss": 1.8791, + "step": 4003 + }, + { + "epoch": 0.391015625, + "grad_norm": 0.194699227809906, + "learning_rate": 0.00036264531886417436, + "loss": 1.8592, + "step": 4004 + }, + { + "epoch": 0.39111328125, + "grad_norm": 0.27056989073753357, + "learning_rate": 0.000362579819029372, + "loss": 1.8255, + "step": 4005 + }, + { + "epoch": 0.3912109375, + "grad_norm": 0.2623845338821411, + "learning_rate": 0.0003625143104461218, + "loss": 1.8555, + "step": 4006 + }, + { + "epoch": 0.39130859375, + "grad_norm": 0.2528698146343231, + "learning_rate": 0.0003624487931209672, + "loss": 1.8176, + "step": 4007 + }, + { + "epoch": 0.39140625, + "grad_norm": 0.2843823730945587, + "learning_rate": 0.000362383267060453, + "loss": 1.8409, + "step": 4008 + }, + { + "epoch": 0.39150390625, + "grad_norm": 0.196132093667984, + "learning_rate": 0.0003623177322711244, + "loss": 1.8331, + "step": 4009 + }, + { + "epoch": 0.3916015625, + "grad_norm": 0.2398262619972229, + "learning_rate": 0.000362252188759528, + "loss": 1.867, + "step": 4010 + }, + { + "epoch": 0.39169921875, + "grad_norm": 0.21861997246742249, + "learning_rate": 0.0003621866365322109, + "loss": 1.869, + "step": 4011 + }, + { + "epoch": 0.391796875, + "grad_norm": 0.23221160471439362, + "learning_rate": 0.0003621210755957212, + "loss": 1.8816, + "step": 4012 + }, + { + "epoch": 0.39189453125, + "grad_norm": 0.18605566024780273, + "learning_rate": 0.0003620555059566079, + "loss": 1.8516, + "step": 4013 + }, + { + "epoch": 0.3919921875, + "grad_norm": 0.2481876164674759, + "learning_rate": 0.0003619899276214208, + "loss": 1.867, + "step": 4014 + }, + { + "epoch": 0.39208984375, + "grad_norm": 0.22675789892673492, + "learning_rate": 0.0003619243405967106, + "loss": 1.8357, + "step": 4015 + }, + { + "epoch": 0.3921875, + "grad_norm": 0.19097070395946503, + "learning_rate": 0.0003618587448890289, + "loss": 1.8706, + "step": 4016 + }, + { + "epoch": 0.39228515625, + "grad_norm": 0.22903457283973694, + "learning_rate": 0.00036179314050492796, + "loss": 1.8436, + "step": 4017 + }, + { + "epoch": 0.3923828125, + "grad_norm": 0.20662012696266174, + "learning_rate": 0.0003617275274509613, + "loss": 1.8151, + "step": 4018 + }, + { + "epoch": 0.39248046875, + "grad_norm": 0.21021722257137299, + "learning_rate": 0.0003616619057336829, + "loss": 1.8779, + "step": 4019 + }, + { + "epoch": 0.392578125, + "grad_norm": 0.22491545975208282, + "learning_rate": 0.00036159627535964795, + "loss": 1.8575, + "step": 4020 + }, + { + "epoch": 0.39267578125, + "grad_norm": 0.24279853701591492, + "learning_rate": 0.0003615306363354122, + "loss": 1.8671, + "step": 4021 + }, + { + "epoch": 0.3927734375, + "grad_norm": 0.23732434213161469, + "learning_rate": 0.0003614649886675326, + "loss": 1.8306, + "step": 4022 + }, + { + "epoch": 0.39287109375, + "grad_norm": 0.17786726355552673, + "learning_rate": 0.0003613993323625665, + "loss": 1.8064, + "step": 4023 + }, + { + "epoch": 0.39296875, + "grad_norm": 0.18526485562324524, + "learning_rate": 0.00036133366742707257, + "loss": 1.8427, + "step": 4024 + }, + { + "epoch": 0.39306640625, + "grad_norm": 0.1718062311410904, + "learning_rate": 0.00036126799386761025, + "loss": 1.8645, + "step": 4025 + }, + { + "epoch": 0.3931640625, + "grad_norm": 0.19269327819347382, + "learning_rate": 0.00036120231169073944, + "loss": 1.8668, + "step": 4026 + }, + { + "epoch": 0.39326171875, + "grad_norm": 0.19223515689373016, + "learning_rate": 0.0003611366209030214, + "loss": 1.861, + "step": 4027 + }, + { + "epoch": 0.393359375, + "grad_norm": 0.18523992598056793, + "learning_rate": 0.000361070921511018, + "loss": 1.8786, + "step": 4028 + }, + { + "epoch": 0.39345703125, + "grad_norm": 0.18942400813102722, + "learning_rate": 0.00036100521352129214, + "loss": 1.8613, + "step": 4029 + }, + { + "epoch": 0.3935546875, + "grad_norm": 0.20724228024482727, + "learning_rate": 0.00036093949694040734, + "loss": 1.7906, + "step": 4030 + }, + { + "epoch": 0.39365234375, + "grad_norm": 0.21637628972530365, + "learning_rate": 0.0003608737717749282, + "loss": 1.89, + "step": 4031 + }, + { + "epoch": 0.39375, + "grad_norm": 0.1924957036972046, + "learning_rate": 0.00036080803803142003, + "loss": 1.8543, + "step": 4032 + }, + { + "epoch": 0.39384765625, + "grad_norm": 0.19536274671554565, + "learning_rate": 0.00036074229571644894, + "loss": 1.8353, + "step": 4033 + }, + { + "epoch": 0.3939453125, + "grad_norm": 0.21658040583133698, + "learning_rate": 0.0003606765448365823, + "loss": 1.8187, + "step": 4034 + }, + { + "epoch": 0.39404296875, + "grad_norm": 0.20503970980644226, + "learning_rate": 0.0003606107853983877, + "loss": 1.8199, + "step": 4035 + }, + { + "epoch": 0.394140625, + "grad_norm": 0.22070644795894623, + "learning_rate": 0.00036054501740843416, + "loss": 1.8097, + "step": 4036 + }, + { + "epoch": 0.39423828125, + "grad_norm": 0.19360551238059998, + "learning_rate": 0.0003604792408732911, + "loss": 1.8259, + "step": 4037 + }, + { + "epoch": 0.3943359375, + "grad_norm": 0.18369770050048828, + "learning_rate": 0.0003604134557995294, + "loss": 1.859, + "step": 4038 + }, + { + "epoch": 0.39443359375, + "grad_norm": 0.219722718000412, + "learning_rate": 0.00036034766219372, + "loss": 1.8831, + "step": 4039 + }, + { + "epoch": 0.39453125, + "grad_norm": 0.18859165906906128, + "learning_rate": 0.0003602818600624353, + "loss": 1.871, + "step": 4040 + }, + { + "epoch": 0.39462890625, + "grad_norm": 0.16907556354999542, + "learning_rate": 0.00036021604941224834, + "loss": 1.8429, + "step": 4041 + }, + { + "epoch": 0.3947265625, + "grad_norm": 0.20368745923042297, + "learning_rate": 0.0003601502302497329, + "loss": 1.8453, + "step": 4042 + }, + { + "epoch": 0.39482421875, + "grad_norm": 0.21178042888641357, + "learning_rate": 0.0003600844025814639, + "loss": 1.8394, + "step": 4043 + }, + { + "epoch": 0.394921875, + "grad_norm": 0.2171480804681778, + "learning_rate": 0.0003600185664140168, + "loss": 1.873, + "step": 4044 + }, + { + "epoch": 0.39501953125, + "grad_norm": 0.2801247239112854, + "learning_rate": 0.0003599527217539682, + "loss": 1.892, + "step": 4045 + }, + { + "epoch": 0.3951171875, + "grad_norm": 0.20998451113700867, + "learning_rate": 0.00035988686860789526, + "loss": 1.8469, + "step": 4046 + }, + { + "epoch": 0.39521484375, + "grad_norm": 0.23407518863677979, + "learning_rate": 0.0003598210069823761, + "loss": 1.8476, + "step": 4047 + }, + { + "epoch": 0.3953125, + "grad_norm": 0.27358564734458923, + "learning_rate": 0.00035975513688398994, + "loss": 1.8614, + "step": 4048 + }, + { + "epoch": 0.39541015625, + "grad_norm": 0.25694334506988525, + "learning_rate": 0.0003596892583193164, + "loss": 1.8174, + "step": 4049 + }, + { + "epoch": 0.3955078125, + "grad_norm": 0.24655091762542725, + "learning_rate": 0.0003596233712949362, + "loss": 1.8475, + "step": 4050 + }, + { + "epoch": 0.39560546875, + "grad_norm": 0.23054777085781097, + "learning_rate": 0.0003595574758174309, + "loss": 1.7855, + "step": 4051 + }, + { + "epoch": 0.395703125, + "grad_norm": 0.31576693058013916, + "learning_rate": 0.00035949157189338294, + "loss": 1.8487, + "step": 4052 + }, + { + "epoch": 0.39580078125, + "grad_norm": 0.21702778339385986, + "learning_rate": 0.00035942565952937547, + "loss": 1.8384, + "step": 4053 + }, + { + "epoch": 0.3958984375, + "grad_norm": 0.2344021499156952, + "learning_rate": 0.0003593597387319925, + "loss": 1.8384, + "step": 4054 + }, + { + "epoch": 0.39599609375, + "grad_norm": 0.2818906903266907, + "learning_rate": 0.000359293809507819, + "loss": 1.8767, + "step": 4055 + }, + { + "epoch": 0.39609375, + "grad_norm": 0.24886588752269745, + "learning_rate": 0.00035922787186344076, + "loss": 1.8564, + "step": 4056 + }, + { + "epoch": 0.39619140625, + "grad_norm": 0.21671739220619202, + "learning_rate": 0.00035916192580544434, + "loss": 1.8982, + "step": 4057 + }, + { + "epoch": 0.3962890625, + "grad_norm": 0.21123698353767395, + "learning_rate": 0.000359095971340417, + "loss": 1.924, + "step": 4058 + }, + { + "epoch": 0.39638671875, + "grad_norm": 0.21664677560329437, + "learning_rate": 0.00035903000847494724, + "loss": 1.8998, + "step": 4059 + }, + { + "epoch": 0.396484375, + "grad_norm": 0.19823968410491943, + "learning_rate": 0.00035896403721562406, + "loss": 1.8309, + "step": 4060 + }, + { + "epoch": 0.39658203125, + "grad_norm": 0.22748810052871704, + "learning_rate": 0.0003588980575690374, + "loss": 1.8967, + "step": 4061 + }, + { + "epoch": 0.3966796875, + "grad_norm": 0.23281057178974152, + "learning_rate": 0.0003588320695417781, + "loss": 1.8736, + "step": 4062 + }, + { + "epoch": 0.39677734375, + "grad_norm": 0.22981210052967072, + "learning_rate": 0.00035876607314043766, + "loss": 1.8353, + "step": 4063 + }, + { + "epoch": 0.396875, + "grad_norm": 0.22894325852394104, + "learning_rate": 0.00035870006837160866, + "loss": 1.8416, + "step": 4064 + }, + { + "epoch": 0.39697265625, + "grad_norm": 0.22248433530330658, + "learning_rate": 0.0003586340552418843, + "loss": 1.8807, + "step": 4065 + }, + { + "epoch": 0.3970703125, + "grad_norm": 0.237218976020813, + "learning_rate": 0.00035856803375785884, + "loss": 1.7974, + "step": 4066 + }, + { + "epoch": 0.39716796875, + "grad_norm": 0.20639021694660187, + "learning_rate": 0.00035850200392612713, + "loss": 1.8686, + "step": 4067 + }, + { + "epoch": 0.397265625, + "grad_norm": 0.21953007578849792, + "learning_rate": 0.00035843596575328495, + "loss": 1.8426, + "step": 4068 + }, + { + "epoch": 0.39736328125, + "grad_norm": 0.21375709772109985, + "learning_rate": 0.00035836991924592903, + "loss": 1.8502, + "step": 4069 + }, + { + "epoch": 0.3974609375, + "grad_norm": 0.26015183329582214, + "learning_rate": 0.0003583038644106567, + "loss": 1.8387, + "step": 4070 + }, + { + "epoch": 0.39755859375, + "grad_norm": 0.22684982419013977, + "learning_rate": 0.00035823780125406637, + "loss": 1.8188, + "step": 4071 + }, + { + "epoch": 0.39765625, + "grad_norm": 0.1907491832971573, + "learning_rate": 0.0003581717297827571, + "loss": 1.8611, + "step": 4072 + }, + { + "epoch": 0.39775390625, + "grad_norm": 0.2399233877658844, + "learning_rate": 0.0003581056500033289, + "loss": 1.8722, + "step": 4073 + }, + { + "epoch": 0.3978515625, + "grad_norm": 0.2267402708530426, + "learning_rate": 0.0003580395619223824, + "loss": 1.8556, + "step": 4074 + }, + { + "epoch": 0.39794921875, + "grad_norm": 0.2641052007675171, + "learning_rate": 0.0003579734655465194, + "loss": 1.8526, + "step": 4075 + }, + { + "epoch": 0.398046875, + "grad_norm": 0.21591079235076904, + "learning_rate": 0.00035790736088234227, + "loss": 1.8107, + "step": 4076 + }, + { + "epoch": 0.39814453125, + "grad_norm": 0.23522786796092987, + "learning_rate": 0.0003578412479364543, + "loss": 1.8761, + "step": 4077 + }, + { + "epoch": 0.3982421875, + "grad_norm": 0.21411925554275513, + "learning_rate": 0.00035777512671545953, + "loss": 1.8791, + "step": 4078 + }, + { + "epoch": 0.39833984375, + "grad_norm": 0.21472613513469696, + "learning_rate": 0.000357708997225963, + "loss": 1.8194, + "step": 4079 + }, + { + "epoch": 0.3984375, + "grad_norm": 0.23808172345161438, + "learning_rate": 0.0003576428594745703, + "loss": 1.8432, + "step": 4080 + }, + { + "epoch": 0.39853515625, + "grad_norm": 0.21339382231235504, + "learning_rate": 0.00035757671346788803, + "loss": 1.8387, + "step": 4081 + }, + { + "epoch": 0.3986328125, + "grad_norm": 0.22620201110839844, + "learning_rate": 0.0003575105592125238, + "loss": 1.8393, + "step": 4082 + }, + { + "epoch": 0.39873046875, + "grad_norm": 0.25172725319862366, + "learning_rate": 0.0003574443967150856, + "loss": 1.8487, + "step": 4083 + }, + { + "epoch": 0.398828125, + "grad_norm": 0.20012016594409943, + "learning_rate": 0.00035737822598218256, + "loss": 1.8415, + "step": 4084 + }, + { + "epoch": 0.39892578125, + "grad_norm": 0.21950571238994598, + "learning_rate": 0.0003573120470204246, + "loss": 1.8832, + "step": 4085 + }, + { + "epoch": 0.3990234375, + "grad_norm": 0.20533329248428345, + "learning_rate": 0.00035724585983642234, + "loss": 1.8099, + "step": 4086 + }, + { + "epoch": 0.39912109375, + "grad_norm": 0.231434166431427, + "learning_rate": 0.0003571796644367873, + "loss": 1.8297, + "step": 4087 + }, + { + "epoch": 0.39921875, + "grad_norm": 0.20492421090602875, + "learning_rate": 0.00035711346082813183, + "loss": 1.8369, + "step": 4088 + }, + { + "epoch": 0.39931640625, + "grad_norm": 0.23543506860733032, + "learning_rate": 0.00035704724901706905, + "loss": 1.8491, + "step": 4089 + }, + { + "epoch": 0.3994140625, + "grad_norm": 0.227260559797287, + "learning_rate": 0.00035698102901021304, + "loss": 1.8817, + "step": 4090 + }, + { + "epoch": 0.39951171875, + "grad_norm": 0.22218318283557892, + "learning_rate": 0.00035691480081417843, + "loss": 1.8093, + "step": 4091 + }, + { + "epoch": 0.399609375, + "grad_norm": 0.20409689843654633, + "learning_rate": 0.00035684856443558095, + "loss": 1.8418, + "step": 4092 + }, + { + "epoch": 0.39970703125, + "grad_norm": 0.21179689466953278, + "learning_rate": 0.000356782319881037, + "loss": 1.8685, + "step": 4093 + }, + { + "epoch": 0.3998046875, + "grad_norm": 0.20115849375724792, + "learning_rate": 0.0003567160671571639, + "loss": 1.8759, + "step": 4094 + }, + { + "epoch": 0.39990234375, + "grad_norm": 0.23569095134735107, + "learning_rate": 0.00035664980627057955, + "loss": 1.8217, + "step": 4095 + }, + { + "epoch": 0.4, + "grad_norm": 0.21255749464035034, + "learning_rate": 0.0003565835372279029, + "loss": 1.8576, + "step": 4096 + }, + { + "epoch": 0.40009765625, + "grad_norm": 0.20328132808208466, + "learning_rate": 0.0003565172600357537, + "loss": 1.8178, + "step": 4097 + }, + { + "epoch": 0.4001953125, + "grad_norm": 0.18217550218105316, + "learning_rate": 0.0003564509747007523, + "loss": 1.8567, + "step": 4098 + }, + { + "epoch": 0.40029296875, + "grad_norm": 0.24595117568969727, + "learning_rate": 0.0003563846812295202, + "loss": 1.8359, + "step": 4099 + }, + { + "epoch": 0.400390625, + "grad_norm": 0.22450943291187286, + "learning_rate": 0.0003563183796286794, + "loss": 1.8305, + "step": 4100 + }, + { + "epoch": 0.40048828125, + "grad_norm": 0.23122942447662354, + "learning_rate": 0.00035625206990485295, + "loss": 1.8603, + "step": 4101 + }, + { + "epoch": 0.4005859375, + "grad_norm": 0.19824737310409546, + "learning_rate": 0.0003561857520646645, + "loss": 1.8617, + "step": 4102 + }, + { + "epoch": 0.40068359375, + "grad_norm": 0.18395096063613892, + "learning_rate": 0.0003561194261147386, + "loss": 1.8598, + "step": 4103 + }, + { + "epoch": 0.40078125, + "grad_norm": 0.2111394703388214, + "learning_rate": 0.0003560530920617009, + "loss": 1.8488, + "step": 4104 + }, + { + "epoch": 0.40087890625, + "grad_norm": 0.20235751569271088, + "learning_rate": 0.00035598674991217715, + "loss": 1.8167, + "step": 4105 + }, + { + "epoch": 0.4009765625, + "grad_norm": 0.20254381000995636, + "learning_rate": 0.00035592039967279466, + "loss": 1.8727, + "step": 4106 + }, + { + "epoch": 0.40107421875, + "grad_norm": 0.19116955995559692, + "learning_rate": 0.0003558540413501811, + "loss": 1.8705, + "step": 4107 + }, + { + "epoch": 0.401171875, + "grad_norm": 0.2449391633272171, + "learning_rate": 0.00035578767495096517, + "loss": 1.8497, + "step": 4108 + }, + { + "epoch": 0.40126953125, + "grad_norm": 0.20743444561958313, + "learning_rate": 0.00035572130048177616, + "loss": 1.8824, + "step": 4109 + }, + { + "epoch": 0.4013671875, + "grad_norm": 0.2034366875886917, + "learning_rate": 0.00035565491794924435, + "loss": 1.8859, + "step": 4110 + }, + { + "epoch": 0.40146484375, + "grad_norm": 0.19879882037639618, + "learning_rate": 0.0003555885273600009, + "loss": 1.8771, + "step": 4111 + }, + { + "epoch": 0.4015625, + "grad_norm": 0.1877567619085312, + "learning_rate": 0.00035552212872067733, + "loss": 1.8553, + "step": 4112 + }, + { + "epoch": 0.40166015625, + "grad_norm": 0.20307008922100067, + "learning_rate": 0.00035545572203790653, + "loss": 1.8379, + "step": 4113 + }, + { + "epoch": 0.4017578125, + "grad_norm": 0.1797867864370346, + "learning_rate": 0.0003553893073183219, + "loss": 1.8359, + "step": 4114 + }, + { + "epoch": 0.40185546875, + "grad_norm": 0.1826634407043457, + "learning_rate": 0.0003553228845685577, + "loss": 1.9227, + "step": 4115 + }, + { + "epoch": 0.401953125, + "grad_norm": 0.17619703710079193, + "learning_rate": 0.0003552564537952489, + "loss": 1.8275, + "step": 4116 + }, + { + "epoch": 0.40205078125, + "grad_norm": 0.1993313729763031, + "learning_rate": 0.0003551900150050313, + "loss": 1.87, + "step": 4117 + }, + { + "epoch": 0.4021484375, + "grad_norm": 0.19552871584892273, + "learning_rate": 0.00035512356820454173, + "loss": 1.8668, + "step": 4118 + }, + { + "epoch": 0.40224609375, + "grad_norm": 0.21230106055736542, + "learning_rate": 0.00035505711340041746, + "loss": 1.8461, + "step": 4119 + }, + { + "epoch": 0.40234375, + "grad_norm": 0.22370152175426483, + "learning_rate": 0.0003549906505992968, + "loss": 1.8724, + "step": 4120 + }, + { + "epoch": 0.40244140625, + "grad_norm": 0.20677858591079712, + "learning_rate": 0.00035492417980781876, + "loss": 1.8431, + "step": 4121 + }, + { + "epoch": 0.4025390625, + "grad_norm": 0.2558901906013489, + "learning_rate": 0.0003548577010326233, + "loss": 1.8964, + "step": 4122 + }, + { + "epoch": 0.40263671875, + "grad_norm": 0.27261173725128174, + "learning_rate": 0.0003547912142803509, + "loss": 1.886, + "step": 4123 + }, + { + "epoch": 0.402734375, + "grad_norm": 0.1867993026971817, + "learning_rate": 0.0003547247195576432, + "loss": 1.8509, + "step": 4124 + }, + { + "epoch": 0.40283203125, + "grad_norm": 0.2584269642829895, + "learning_rate": 0.00035465821687114224, + "loss": 1.87, + "step": 4125 + }, + { + "epoch": 0.4029296875, + "grad_norm": 0.2057262659072876, + "learning_rate": 0.00035459170622749117, + "loss": 1.8496, + "step": 4126 + }, + { + "epoch": 0.40302734375, + "grad_norm": 0.22833289206027985, + "learning_rate": 0.0003545251876333337, + "loss": 1.8487, + "step": 4127 + }, + { + "epoch": 0.403125, + "grad_norm": 0.21056689321994781, + "learning_rate": 0.00035445866109531455, + "loss": 1.849, + "step": 4128 + }, + { + "epoch": 0.40322265625, + "grad_norm": 0.18776793777942657, + "learning_rate": 0.0003543921266200791, + "loss": 1.8724, + "step": 4129 + }, + { + "epoch": 0.4033203125, + "grad_norm": 0.2139902561903, + "learning_rate": 0.0003543255842142736, + "loss": 1.8522, + "step": 4130 + }, + { + "epoch": 0.40341796875, + "grad_norm": 0.2051902860403061, + "learning_rate": 0.0003542590338845449, + "loss": 1.8631, + "step": 4131 + }, + { + "epoch": 0.403515625, + "grad_norm": 0.21415852010250092, + "learning_rate": 0.00035419247563754094, + "loss": 1.8147, + "step": 4132 + }, + { + "epoch": 0.40361328125, + "grad_norm": 0.19563965499401093, + "learning_rate": 0.00035412590947991017, + "loss": 1.81, + "step": 4133 + }, + { + "epoch": 0.4037109375, + "grad_norm": 0.17605456709861755, + "learning_rate": 0.0003540593354183022, + "loss": 1.8285, + "step": 4134 + }, + { + "epoch": 0.40380859375, + "grad_norm": 0.24080799520015717, + "learning_rate": 0.0003539927534593668, + "loss": 1.8393, + "step": 4135 + }, + { + "epoch": 0.40390625, + "grad_norm": 0.24760200083255768, + "learning_rate": 0.0003539261636097553, + "loss": 1.8982, + "step": 4136 + }, + { + "epoch": 0.40400390625, + "grad_norm": 0.21751075983047485, + "learning_rate": 0.0003538595658761192, + "loss": 1.8246, + "step": 4137 + }, + { + "epoch": 0.4041015625, + "grad_norm": 0.19909746944904327, + "learning_rate": 0.00035379296026511115, + "loss": 1.8526, + "step": 4138 + }, + { + "epoch": 0.40419921875, + "grad_norm": 0.24126000702381134, + "learning_rate": 0.0003537263467833845, + "loss": 1.865, + "step": 4139 + }, + { + "epoch": 0.404296875, + "grad_norm": 0.22304266691207886, + "learning_rate": 0.0003536597254375931, + "loss": 1.8456, + "step": 4140 + }, + { + "epoch": 0.40439453125, + "grad_norm": 0.20051011443138123, + "learning_rate": 0.0003535930962343921, + "loss": 1.8121, + "step": 4141 + }, + { + "epoch": 0.4044921875, + "grad_norm": 0.2357867807149887, + "learning_rate": 0.00035352645918043695, + "loss": 1.8646, + "step": 4142 + }, + { + "epoch": 0.40458984375, + "grad_norm": 0.25244826078414917, + "learning_rate": 0.00035345981428238434, + "loss": 1.892, + "step": 4143 + }, + { + "epoch": 0.4046875, + "grad_norm": 0.2515060603618622, + "learning_rate": 0.0003533931615468913, + "loss": 1.8368, + "step": 4144 + }, + { + "epoch": 0.40478515625, + "grad_norm": 0.2245102971792221, + "learning_rate": 0.00035332650098061593, + "loss": 1.8444, + "step": 4145 + }, + { + "epoch": 0.4048828125, + "grad_norm": 0.3059788644313812, + "learning_rate": 0.00035325983259021707, + "loss": 1.8445, + "step": 4146 + }, + { + "epoch": 0.40498046875, + "grad_norm": 0.2570042610168457, + "learning_rate": 0.0003531931563823542, + "loss": 1.8501, + "step": 4147 + }, + { + "epoch": 0.405078125, + "grad_norm": 0.23951447010040283, + "learning_rate": 0.0003531264723636877, + "loss": 1.839, + "step": 4148 + }, + { + "epoch": 0.40517578125, + "grad_norm": 0.28501424193382263, + "learning_rate": 0.00035305978054087886, + "loss": 1.823, + "step": 4149 + }, + { + "epoch": 0.4052734375, + "grad_norm": 0.21223093569278717, + "learning_rate": 0.0003529930809205894, + "loss": 1.8349, + "step": 4150 + }, + { + "epoch": 0.40537109375, + "grad_norm": 0.25295740365982056, + "learning_rate": 0.00035292637350948207, + "loss": 1.8081, + "step": 4151 + }, + { + "epoch": 0.40546875, + "grad_norm": 0.2550055980682373, + "learning_rate": 0.00035285965831422054, + "loss": 1.8832, + "step": 4152 + }, + { + "epoch": 0.40556640625, + "grad_norm": 0.2280154526233673, + "learning_rate": 0.00035279293534146877, + "loss": 1.841, + "step": 4153 + }, + { + "epoch": 0.4056640625, + "grad_norm": 0.20514193177223206, + "learning_rate": 0.00035272620459789205, + "loss": 1.861, + "step": 4154 + }, + { + "epoch": 0.40576171875, + "grad_norm": 0.2677548825740814, + "learning_rate": 0.000352659466090156, + "loss": 1.8526, + "step": 4155 + }, + { + "epoch": 0.405859375, + "grad_norm": 0.23861974477767944, + "learning_rate": 0.00035259271982492735, + "loss": 1.8089, + "step": 4156 + }, + { + "epoch": 0.40595703125, + "grad_norm": 0.22723138332366943, + "learning_rate": 0.00035252596580887343, + "loss": 1.8546, + "step": 4157 + }, + { + "epoch": 0.4060546875, + "grad_norm": 0.2531287670135498, + "learning_rate": 0.0003524592040486622, + "loss": 1.8654, + "step": 4158 + }, + { + "epoch": 0.40615234375, + "grad_norm": 0.2574120759963989, + "learning_rate": 0.00035239243455096285, + "loss": 1.8565, + "step": 4159 + }, + { + "epoch": 0.40625, + "grad_norm": 0.20986592769622803, + "learning_rate": 0.0003523256573224449, + "loss": 1.836, + "step": 4160 + }, + { + "epoch": 0.40634765625, + "grad_norm": 0.1971091628074646, + "learning_rate": 0.0003522588723697789, + "loss": 1.8979, + "step": 4161 + }, + { + "epoch": 0.4064453125, + "grad_norm": 0.2572074234485626, + "learning_rate": 0.00035219207969963587, + "loss": 1.8226, + "step": 4162 + }, + { + "epoch": 0.40654296875, + "grad_norm": 0.17880785465240479, + "learning_rate": 0.0003521252793186879, + "loss": 1.797, + "step": 4163 + }, + { + "epoch": 0.406640625, + "grad_norm": 0.23887290060520172, + "learning_rate": 0.00035205847123360794, + "loss": 1.8581, + "step": 4164 + }, + { + "epoch": 0.40673828125, + "grad_norm": 0.21064692735671997, + "learning_rate": 0.00035199165545106925, + "loss": 1.871, + "step": 4165 + }, + { + "epoch": 0.4068359375, + "grad_norm": 0.19952860474586487, + "learning_rate": 0.00035192483197774634, + "loss": 1.8349, + "step": 4166 + }, + { + "epoch": 0.40693359375, + "grad_norm": 0.19057486951351166, + "learning_rate": 0.0003518580008203142, + "loss": 1.8081, + "step": 4167 + }, + { + "epoch": 0.40703125, + "grad_norm": 0.19488559663295746, + "learning_rate": 0.00035179116198544877, + "loss": 1.8361, + "step": 4168 + }, + { + "epoch": 0.40712890625, + "grad_norm": 0.18777482211589813, + "learning_rate": 0.00035172431547982644, + "loss": 1.8531, + "step": 4169 + }, + { + "epoch": 0.4072265625, + "grad_norm": 0.2459339201450348, + "learning_rate": 0.0003516574613101247, + "loss": 1.8105, + "step": 4170 + }, + { + "epoch": 0.40732421875, + "grad_norm": 0.19164249300956726, + "learning_rate": 0.0003515905994830218, + "loss": 1.8585, + "step": 4171 + }, + { + "epoch": 0.407421875, + "grad_norm": 0.2304055094718933, + "learning_rate": 0.00035152373000519644, + "loss": 1.7608, + "step": 4172 + }, + { + "epoch": 0.40751953125, + "grad_norm": 0.22517216205596924, + "learning_rate": 0.00035145685288332846, + "loss": 1.8473, + "step": 4173 + }, + { + "epoch": 0.4076171875, + "grad_norm": 0.20082591474056244, + "learning_rate": 0.0003513899681240981, + "loss": 1.8695, + "step": 4174 + }, + { + "epoch": 0.40771484375, + "grad_norm": 0.22921518981456757, + "learning_rate": 0.0003513230757341869, + "loss": 1.8218, + "step": 4175 + }, + { + "epoch": 0.4078125, + "grad_norm": 0.19029761850833893, + "learning_rate": 0.0003512561757202764, + "loss": 1.8311, + "step": 4176 + }, + { + "epoch": 0.40791015625, + "grad_norm": 0.19780279695987701, + "learning_rate": 0.0003511892680890496, + "loss": 1.8341, + "step": 4177 + }, + { + "epoch": 0.4080078125, + "grad_norm": 0.19720710813999176, + "learning_rate": 0.00035112235284718993, + "loss": 1.8473, + "step": 4178 + }, + { + "epoch": 0.40810546875, + "grad_norm": 0.19246746599674225, + "learning_rate": 0.00035105543000138147, + "loss": 1.8593, + "step": 4179 + }, + { + "epoch": 0.408203125, + "grad_norm": 0.19287921488285065, + "learning_rate": 0.00035098849955830943, + "loss": 1.9073, + "step": 4180 + }, + { + "epoch": 0.40830078125, + "grad_norm": 0.22399508953094482, + "learning_rate": 0.0003509215615246595, + "loss": 1.8363, + "step": 4181 + }, + { + "epoch": 0.4083984375, + "grad_norm": 0.2298484742641449, + "learning_rate": 0.00035085461590711817, + "loss": 1.8709, + "step": 4182 + }, + { + "epoch": 0.40849609375, + "grad_norm": 0.18680132925510406, + "learning_rate": 0.0003507876627123727, + "loss": 1.8364, + "step": 4183 + }, + { + "epoch": 0.40859375, + "grad_norm": 0.21050740778446198, + "learning_rate": 0.00035072070194711106, + "loss": 1.8382, + "step": 4184 + }, + { + "epoch": 0.40869140625, + "grad_norm": 0.24991776049137115, + "learning_rate": 0.00035065373361802227, + "loss": 1.8289, + "step": 4185 + }, + { + "epoch": 0.4087890625, + "grad_norm": 0.22001494467258453, + "learning_rate": 0.00035058675773179566, + "loss": 1.8253, + "step": 4186 + }, + { + "epoch": 0.40888671875, + "grad_norm": 0.23366527259349823, + "learning_rate": 0.0003505197742951216, + "loss": 1.8731, + "step": 4187 + }, + { + "epoch": 0.408984375, + "grad_norm": 0.1992902010679245, + "learning_rate": 0.0003504527833146911, + "loss": 1.8637, + "step": 4188 + }, + { + "epoch": 0.40908203125, + "grad_norm": 0.18898622691631317, + "learning_rate": 0.000350385784797196, + "loss": 1.8352, + "step": 4189 + }, + { + "epoch": 0.4091796875, + "grad_norm": 0.24903878569602966, + "learning_rate": 0.00035031877874932893, + "loss": 1.8385, + "step": 4190 + }, + { + "epoch": 0.40927734375, + "grad_norm": 0.20514540374279022, + "learning_rate": 0.00035025176517778305, + "loss": 1.85, + "step": 4191 + }, + { + "epoch": 0.409375, + "grad_norm": 0.2659608721733093, + "learning_rate": 0.00035018474408925253, + "loss": 1.8314, + "step": 4192 + }, + { + "epoch": 0.40947265625, + "grad_norm": 0.17859287559986115, + "learning_rate": 0.0003501177154904321, + "loss": 1.8438, + "step": 4193 + }, + { + "epoch": 0.4095703125, + "grad_norm": 0.19967199862003326, + "learning_rate": 0.00035005067938801744, + "loss": 1.8129, + "step": 4194 + }, + { + "epoch": 0.40966796875, + "grad_norm": 0.21078793704509735, + "learning_rate": 0.0003499836357887048, + "loss": 1.8088, + "step": 4195 + }, + { + "epoch": 0.409765625, + "grad_norm": 0.16362448036670685, + "learning_rate": 0.00034991658469919117, + "loss": 1.8544, + "step": 4196 + }, + { + "epoch": 0.40986328125, + "grad_norm": 0.1888558268547058, + "learning_rate": 0.0003498495261261746, + "loss": 1.8535, + "step": 4197 + }, + { + "epoch": 0.4099609375, + "grad_norm": 0.19681403040885925, + "learning_rate": 0.00034978246007635335, + "loss": 1.8286, + "step": 4198 + }, + { + "epoch": 0.41005859375, + "grad_norm": 0.1948511153459549, + "learning_rate": 0.0003497153865564268, + "loss": 1.8237, + "step": 4199 + }, + { + "epoch": 0.41015625, + "grad_norm": 0.21308070421218872, + "learning_rate": 0.00034964830557309513, + "loss": 1.8074, + "step": 4200 + }, + { + "epoch": 0.41025390625, + "grad_norm": 0.20020486414432526, + "learning_rate": 0.000349581217133059, + "loss": 1.8793, + "step": 4201 + }, + { + "epoch": 0.4103515625, + "grad_norm": 0.20001058280467987, + "learning_rate": 0.00034951412124302006, + "loss": 1.8133, + "step": 4202 + }, + { + "epoch": 0.41044921875, + "grad_norm": 0.21066218614578247, + "learning_rate": 0.00034944701790968054, + "loss": 1.8185, + "step": 4203 + }, + { + "epoch": 0.410546875, + "grad_norm": 0.22479777038097382, + "learning_rate": 0.0003493799071397435, + "loss": 1.8591, + "step": 4204 + }, + { + "epoch": 0.41064453125, + "grad_norm": 0.1952837109565735, + "learning_rate": 0.00034931278893991265, + "loss": 1.8423, + "step": 4205 + }, + { + "epoch": 0.4107421875, + "grad_norm": 0.28419220447540283, + "learning_rate": 0.0003492456633168925, + "loss": 1.8809, + "step": 4206 + }, + { + "epoch": 0.41083984375, + "grad_norm": 0.22131340205669403, + "learning_rate": 0.0003491785302773883, + "loss": 1.8141, + "step": 4207 + }, + { + "epoch": 0.4109375, + "grad_norm": 0.2682664096355438, + "learning_rate": 0.0003491113898281062, + "loss": 1.8872, + "step": 4208 + }, + { + "epoch": 0.41103515625, + "grad_norm": 0.24773511290550232, + "learning_rate": 0.0003490442419757528, + "loss": 1.8325, + "step": 4209 + }, + { + "epoch": 0.4111328125, + "grad_norm": 0.2604011595249176, + "learning_rate": 0.0003489770867270356, + "loss": 1.8726, + "step": 4210 + }, + { + "epoch": 0.41123046875, + "grad_norm": 0.2504948377609253, + "learning_rate": 0.00034890992408866285, + "loss": 1.8481, + "step": 4211 + }, + { + "epoch": 0.411328125, + "grad_norm": 0.32314008474349976, + "learning_rate": 0.0003488427540673433, + "loss": 1.8546, + "step": 4212 + }, + { + "epoch": 0.41142578125, + "grad_norm": 0.285114049911499, + "learning_rate": 0.000348775576669787, + "loss": 1.8531, + "step": 4213 + }, + { + "epoch": 0.4115234375, + "grad_norm": 0.25935712456703186, + "learning_rate": 0.000348708391902704, + "loss": 1.8106, + "step": 4214 + }, + { + "epoch": 0.41162109375, + "grad_norm": 0.2512550950050354, + "learning_rate": 0.00034864119977280584, + "loss": 1.8648, + "step": 4215 + }, + { + "epoch": 0.41171875, + "grad_norm": 0.333379864692688, + "learning_rate": 0.00034857400028680415, + "loss": 1.841, + "step": 4216 + }, + { + "epoch": 0.41181640625, + "grad_norm": 0.25787755846977234, + "learning_rate": 0.0003485067934514116, + "loss": 1.869, + "step": 4217 + }, + { + "epoch": 0.4119140625, + "grad_norm": 0.19393840432167053, + "learning_rate": 0.0003484395792733416, + "loss": 1.838, + "step": 4218 + }, + { + "epoch": 0.41201171875, + "grad_norm": 0.2149869054555893, + "learning_rate": 0.0003483723577593083, + "loss": 1.8729, + "step": 4219 + }, + { + "epoch": 0.412109375, + "grad_norm": 0.23125958442687988, + "learning_rate": 0.0003483051289160265, + "loss": 1.837, + "step": 4220 + }, + { + "epoch": 0.41220703125, + "grad_norm": 0.1976689249277115, + "learning_rate": 0.00034823789275021164, + "loss": 1.8447, + "step": 4221 + }, + { + "epoch": 0.4123046875, + "grad_norm": 0.24214595556259155, + "learning_rate": 0.00034817064926858017, + "loss": 1.8098, + "step": 4222 + }, + { + "epoch": 0.41240234375, + "grad_norm": 0.22756950557231903, + "learning_rate": 0.00034810339847784913, + "loss": 1.8572, + "step": 4223 + }, + { + "epoch": 0.4125, + "grad_norm": 0.20165514945983887, + "learning_rate": 0.0003480361403847361, + "loss": 1.8555, + "step": 4224 + }, + { + "epoch": 0.41259765625, + "grad_norm": 0.27508628368377686, + "learning_rate": 0.0003479688749959598, + "loss": 1.8739, + "step": 4225 + }, + { + "epoch": 0.4126953125, + "grad_norm": 0.19843104481697083, + "learning_rate": 0.00034790160231823925, + "loss": 1.8419, + "step": 4226 + }, + { + "epoch": 0.41279296875, + "grad_norm": 0.20928111672401428, + "learning_rate": 0.0003478343223582946, + "loss": 1.8809, + "step": 4227 + }, + { + "epoch": 0.412890625, + "grad_norm": 0.21286095678806305, + "learning_rate": 0.0003477670351228462, + "loss": 1.8071, + "step": 4228 + }, + { + "epoch": 0.41298828125, + "grad_norm": 0.183101087808609, + "learning_rate": 0.0003476997406186158, + "loss": 1.836, + "step": 4229 + }, + { + "epoch": 0.4130859375, + "grad_norm": 0.16599592566490173, + "learning_rate": 0.00034763243885232536, + "loss": 1.8283, + "step": 4230 + }, + { + "epoch": 0.41318359375, + "grad_norm": 0.2242308109998703, + "learning_rate": 0.00034756512983069777, + "loss": 1.8075, + "step": 4231 + }, + { + "epoch": 0.41328125, + "grad_norm": 0.19931399822235107, + "learning_rate": 0.0003474978135604565, + "loss": 1.8351, + "step": 4232 + }, + { + "epoch": 0.41337890625, + "grad_norm": 0.20005737245082855, + "learning_rate": 0.000347430490048326, + "loss": 1.8304, + "step": 4233 + }, + { + "epoch": 0.4134765625, + "grad_norm": 0.23433679342269897, + "learning_rate": 0.00034736315930103124, + "loss": 1.861, + "step": 4234 + }, + { + "epoch": 0.41357421875, + "grad_norm": 0.2258288711309433, + "learning_rate": 0.0003472958213252978, + "loss": 1.8316, + "step": 4235 + }, + { + "epoch": 0.413671875, + "grad_norm": 0.19864246249198914, + "learning_rate": 0.0003472284761278524, + "loss": 1.8173, + "step": 4236 + }, + { + "epoch": 0.41376953125, + "grad_norm": 0.2385239154100418, + "learning_rate": 0.00034716112371542215, + "loss": 1.8522, + "step": 4237 + }, + { + "epoch": 0.4138671875, + "grad_norm": 0.2307146191596985, + "learning_rate": 0.00034709376409473484, + "loss": 1.8446, + "step": 4238 + }, + { + "epoch": 0.41396484375, + "grad_norm": 0.22357000410556793, + "learning_rate": 0.0003470263972725193, + "loss": 1.8856, + "step": 4239 + }, + { + "epoch": 0.4140625, + "grad_norm": 0.20348325371742249, + "learning_rate": 0.0003469590232555046, + "loss": 1.8688, + "step": 4240 + }, + { + "epoch": 0.41416015625, + "grad_norm": 0.21446335315704346, + "learning_rate": 0.0003468916420504211, + "loss": 1.8158, + "step": 4241 + }, + { + "epoch": 0.4142578125, + "grad_norm": 0.23739059269428253, + "learning_rate": 0.00034682425366399944, + "loss": 1.8127, + "step": 4242 + }, + { + "epoch": 0.41435546875, + "grad_norm": 0.16460199654102325, + "learning_rate": 0.0003467568581029712, + "loss": 1.83, + "step": 4243 + }, + { + "epoch": 0.414453125, + "grad_norm": 0.22891512513160706, + "learning_rate": 0.0003466894553740685, + "loss": 1.8249, + "step": 4244 + }, + { + "epoch": 0.41455078125, + "grad_norm": 0.21211554110050201, + "learning_rate": 0.00034662204548402425, + "loss": 1.8282, + "step": 4245 + }, + { + "epoch": 0.4146484375, + "grad_norm": 0.23098908364772797, + "learning_rate": 0.00034655462843957225, + "loss": 1.8365, + "step": 4246 + }, + { + "epoch": 0.41474609375, + "grad_norm": 0.2643522024154663, + "learning_rate": 0.0003464872042474468, + "loss": 1.7939, + "step": 4247 + }, + { + "epoch": 0.41484375, + "grad_norm": 0.179979145526886, + "learning_rate": 0.00034641977291438293, + "loss": 1.8654, + "step": 4248 + }, + { + "epoch": 0.41494140625, + "grad_norm": 0.2652702331542969, + "learning_rate": 0.00034635233444711645, + "loss": 1.8313, + "step": 4249 + }, + { + "epoch": 0.4150390625, + "grad_norm": 0.18599368631839752, + "learning_rate": 0.00034628488885238393, + "loss": 1.8835, + "step": 4250 + }, + { + "epoch": 0.41513671875, + "grad_norm": 0.24554285407066345, + "learning_rate": 0.00034621743613692254, + "loss": 1.8121, + "step": 4251 + }, + { + "epoch": 0.415234375, + "grad_norm": 0.2408851832151413, + "learning_rate": 0.00034614997630747006, + "loss": 1.8248, + "step": 4252 + }, + { + "epoch": 0.41533203125, + "grad_norm": 0.2368103414773941, + "learning_rate": 0.00034608250937076545, + "loss": 1.8479, + "step": 4253 + }, + { + "epoch": 0.4154296875, + "grad_norm": 0.20753245055675507, + "learning_rate": 0.0003460150353335479, + "loss": 1.865, + "step": 4254 + }, + { + "epoch": 0.41552734375, + "grad_norm": 0.20561878383159637, + "learning_rate": 0.00034594755420255737, + "loss": 1.863, + "step": 4255 + }, + { + "epoch": 0.415625, + "grad_norm": 0.24068519473075867, + "learning_rate": 0.0003458800659845347, + "loss": 1.8511, + "step": 4256 + }, + { + "epoch": 0.41572265625, + "grad_norm": 0.23239868879318237, + "learning_rate": 0.00034581257068622145, + "loss": 1.9037, + "step": 4257 + }, + { + "epoch": 0.4158203125, + "grad_norm": 0.23503153026103973, + "learning_rate": 0.0003457450683143597, + "loss": 1.7994, + "step": 4258 + }, + { + "epoch": 0.41591796875, + "grad_norm": 0.22664009034633636, + "learning_rate": 0.0003456775588756924, + "loss": 1.8591, + "step": 4259 + }, + { + "epoch": 0.416015625, + "grad_norm": 0.32614830136299133, + "learning_rate": 0.00034561004237696304, + "loss": 1.8632, + "step": 4260 + }, + { + "epoch": 0.41611328125, + "grad_norm": 0.24125008285045624, + "learning_rate": 0.0003455425188249161, + "loss": 1.8563, + "step": 4261 + }, + { + "epoch": 0.4162109375, + "grad_norm": 0.22518505156040192, + "learning_rate": 0.00034547498822629644, + "loss": 1.8402, + "step": 4262 + }, + { + "epoch": 0.41630859375, + "grad_norm": 0.2516859471797943, + "learning_rate": 0.0003454074505878498, + "loss": 1.8494, + "step": 4263 + }, + { + "epoch": 0.41640625, + "grad_norm": 0.2525559067726135, + "learning_rate": 0.00034533990591632254, + "loss": 1.8434, + "step": 4264 + }, + { + "epoch": 0.41650390625, + "grad_norm": 0.16537804901599884, + "learning_rate": 0.00034527235421846195, + "loss": 1.8087, + "step": 4265 + }, + { + "epoch": 0.4166015625, + "grad_norm": 0.2685443162918091, + "learning_rate": 0.0003452047955010156, + "loss": 1.827, + "step": 4266 + }, + { + "epoch": 0.41669921875, + "grad_norm": 0.20350132882595062, + "learning_rate": 0.0003451372297707322, + "loss": 1.852, + "step": 4267 + }, + { + "epoch": 0.416796875, + "grad_norm": 0.21128815412521362, + "learning_rate": 0.00034506965703436093, + "loss": 1.8463, + "step": 4268 + }, + { + "epoch": 0.41689453125, + "grad_norm": 0.20447909832000732, + "learning_rate": 0.00034500207729865165, + "loss": 1.8083, + "step": 4269 + }, + { + "epoch": 0.4169921875, + "grad_norm": 0.20351362228393555, + "learning_rate": 0.00034493449057035504, + "loss": 1.874, + "step": 4270 + }, + { + "epoch": 0.41708984375, + "grad_norm": 0.2231147587299347, + "learning_rate": 0.00034486689685622236, + "loss": 1.8357, + "step": 4271 + }, + { + "epoch": 0.4171875, + "grad_norm": 0.19443069398403168, + "learning_rate": 0.00034479929616300565, + "loss": 1.8425, + "step": 4272 + }, + { + "epoch": 0.41728515625, + "grad_norm": 0.24243175983428955, + "learning_rate": 0.00034473168849745764, + "loss": 1.8664, + "step": 4273 + }, + { + "epoch": 0.4173828125, + "grad_norm": 0.2010633796453476, + "learning_rate": 0.0003446640738663316, + "loss": 1.8539, + "step": 4274 + }, + { + "epoch": 0.41748046875, + "grad_norm": 0.25846266746520996, + "learning_rate": 0.000344596452276382, + "loss": 1.829, + "step": 4275 + }, + { + "epoch": 0.417578125, + "grad_norm": 0.22176463901996613, + "learning_rate": 0.0003445288237343632, + "loss": 1.8324, + "step": 4276 + }, + { + "epoch": 0.41767578125, + "grad_norm": 0.21178513765335083, + "learning_rate": 0.00034446118824703087, + "loss": 1.8216, + "step": 4277 + }, + { + "epoch": 0.4177734375, + "grad_norm": 0.25420916080474854, + "learning_rate": 0.0003443935458211413, + "loss": 1.8528, + "step": 4278 + }, + { + "epoch": 0.41787109375, + "grad_norm": 0.1733928769826889, + "learning_rate": 0.0003443258964634512, + "loss": 1.8562, + "step": 4279 + }, + { + "epoch": 0.41796875, + "grad_norm": 0.22979426383972168, + "learning_rate": 0.00034425824018071826, + "loss": 1.8582, + "step": 4280 + }, + { + "epoch": 0.41806640625, + "grad_norm": 0.18637573719024658, + "learning_rate": 0.0003441905769797007, + "loss": 1.8275, + "step": 4281 + }, + { + "epoch": 0.4181640625, + "grad_norm": 0.18606674671173096, + "learning_rate": 0.00034412290686715747, + "loss": 1.8616, + "step": 4282 + }, + { + "epoch": 0.41826171875, + "grad_norm": 0.18189047276973724, + "learning_rate": 0.00034405522984984815, + "loss": 1.8007, + "step": 4283 + }, + { + "epoch": 0.418359375, + "grad_norm": 0.21638208627700806, + "learning_rate": 0.0003439875459345332, + "loss": 1.8307, + "step": 4284 + }, + { + "epoch": 0.41845703125, + "grad_norm": 0.19453690946102142, + "learning_rate": 0.00034391985512797357, + "loss": 1.8311, + "step": 4285 + }, + { + "epoch": 0.4185546875, + "grad_norm": 0.22609570622444153, + "learning_rate": 0.000343852157436931, + "loss": 1.8486, + "step": 4286 + }, + { + "epoch": 0.41865234375, + "grad_norm": 0.1817261129617691, + "learning_rate": 0.0003437844528681679, + "loss": 1.866, + "step": 4287 + }, + { + "epoch": 0.41875, + "grad_norm": 0.1934969276189804, + "learning_rate": 0.00034371674142844727, + "loss": 1.8448, + "step": 4288 + }, + { + "epoch": 0.41884765625, + "grad_norm": 0.17022395133972168, + "learning_rate": 0.00034364902312453307, + "loss": 1.809, + "step": 4289 + }, + { + "epoch": 0.4189453125, + "grad_norm": 0.221804678440094, + "learning_rate": 0.00034358129796318947, + "loss": 1.8872, + "step": 4290 + }, + { + "epoch": 0.41904296875, + "grad_norm": 0.22451527416706085, + "learning_rate": 0.0003435135659511819, + "loss": 1.8112, + "step": 4291 + }, + { + "epoch": 0.419140625, + "grad_norm": 0.23531721532344818, + "learning_rate": 0.00034344582709527606, + "loss": 1.8643, + "step": 4292 + }, + { + "epoch": 0.41923828125, + "grad_norm": 0.206741601228714, + "learning_rate": 0.00034337808140223844, + "loss": 1.8382, + "step": 4293 + }, + { + "epoch": 0.4193359375, + "grad_norm": 0.26458290219306946, + "learning_rate": 0.0003433103288788362, + "loss": 1.8481, + "step": 4294 + }, + { + "epoch": 0.41943359375, + "grad_norm": 0.2532312870025635, + "learning_rate": 0.0003432425695318373, + "loss": 1.8158, + "step": 4295 + }, + { + "epoch": 0.41953125, + "grad_norm": 0.20990796387195587, + "learning_rate": 0.00034317480336801037, + "loss": 1.8749, + "step": 4296 + }, + { + "epoch": 0.41962890625, + "grad_norm": 0.23719432950019836, + "learning_rate": 0.0003431070303941245, + "loss": 1.8536, + "step": 4297 + }, + { + "epoch": 0.4197265625, + "grad_norm": 0.22180777788162231, + "learning_rate": 0.00034303925061694967, + "loss": 1.8891, + "step": 4298 + }, + { + "epoch": 0.41982421875, + "grad_norm": 0.225361168384552, + "learning_rate": 0.00034297146404325653, + "loss": 1.838, + "step": 4299 + }, + { + "epoch": 0.419921875, + "grad_norm": 0.23712079226970673, + "learning_rate": 0.0003429036706798162, + "loss": 1.8225, + "step": 4300 + }, + { + "epoch": 0.42001953125, + "grad_norm": 0.2806594967842102, + "learning_rate": 0.00034283587053340084, + "loss": 1.8636, + "step": 4301 + }, + { + "epoch": 0.4201171875, + "grad_norm": 0.21580100059509277, + "learning_rate": 0.0003427680636107829, + "loss": 1.834, + "step": 4302 + }, + { + "epoch": 0.42021484375, + "grad_norm": 0.24084138870239258, + "learning_rate": 0.0003427002499187358, + "loss": 1.8283, + "step": 4303 + }, + { + "epoch": 0.4203125, + "grad_norm": 0.26236194372177124, + "learning_rate": 0.00034263242946403356, + "loss": 1.8105, + "step": 4304 + }, + { + "epoch": 0.42041015625, + "grad_norm": 0.24141961336135864, + "learning_rate": 0.0003425646022534508, + "loss": 1.8311, + "step": 4305 + }, + { + "epoch": 0.4205078125, + "grad_norm": 0.17347706854343414, + "learning_rate": 0.0003424967682937627, + "loss": 1.7743, + "step": 4306 + }, + { + "epoch": 0.42060546875, + "grad_norm": 0.27681705355644226, + "learning_rate": 0.0003424289275917455, + "loss": 1.8708, + "step": 4307 + }, + { + "epoch": 0.420703125, + "grad_norm": 0.22059215605258942, + "learning_rate": 0.00034236108015417584, + "loss": 1.809, + "step": 4308 + }, + { + "epoch": 0.42080078125, + "grad_norm": 0.2038775086402893, + "learning_rate": 0.000342293225987831, + "loss": 1.8531, + "step": 4309 + }, + { + "epoch": 0.4208984375, + "grad_norm": 0.1890358030796051, + "learning_rate": 0.0003422253650994891, + "loss": 1.8161, + "step": 4310 + }, + { + "epoch": 0.42099609375, + "grad_norm": 0.22576870024204254, + "learning_rate": 0.00034215749749592873, + "loss": 1.8652, + "step": 4311 + }, + { + "epoch": 0.42109375, + "grad_norm": 0.2194606214761734, + "learning_rate": 0.0003420896231839293, + "loss": 1.7729, + "step": 4312 + }, + { + "epoch": 0.42119140625, + "grad_norm": 0.2838427424430847, + "learning_rate": 0.000342021742170271, + "loss": 1.856, + "step": 4313 + }, + { + "epoch": 0.4212890625, + "grad_norm": 0.23746374249458313, + "learning_rate": 0.0003419538544617342, + "loss": 1.8322, + "step": 4314 + }, + { + "epoch": 0.42138671875, + "grad_norm": 0.1634446680545807, + "learning_rate": 0.00034188596006510066, + "loss": 1.8129, + "step": 4315 + }, + { + "epoch": 0.421484375, + "grad_norm": 0.23567970097064972, + "learning_rate": 0.00034181805898715216, + "loss": 1.8431, + "step": 4316 + }, + { + "epoch": 0.42158203125, + "grad_norm": 0.20782487094402313, + "learning_rate": 0.0003417501512346717, + "loss": 1.8508, + "step": 4317 + }, + { + "epoch": 0.4216796875, + "grad_norm": 0.18531768023967743, + "learning_rate": 0.0003416822368144424, + "loss": 1.8645, + "step": 4318 + }, + { + "epoch": 0.42177734375, + "grad_norm": 0.21057714521884918, + "learning_rate": 0.0003416143157332483, + "loss": 1.8604, + "step": 4319 + }, + { + "epoch": 0.421875, + "grad_norm": 0.1936180740594864, + "learning_rate": 0.0003415463879978743, + "loss": 1.8278, + "step": 4320 + }, + { + "epoch": 0.42197265625, + "grad_norm": 0.20181427896022797, + "learning_rate": 0.0003414784536151056, + "loss": 1.8653, + "step": 4321 + }, + { + "epoch": 0.4220703125, + "grad_norm": 0.21293999254703522, + "learning_rate": 0.00034141051259172845, + "loss": 1.8462, + "step": 4322 + }, + { + "epoch": 0.42216796875, + "grad_norm": 0.18978172540664673, + "learning_rate": 0.0003413425649345293, + "loss": 1.8836, + "step": 4323 + }, + { + "epoch": 0.422265625, + "grad_norm": 0.2115108072757721, + "learning_rate": 0.0003412746106502958, + "loss": 1.8257, + "step": 4324 + }, + { + "epoch": 0.42236328125, + "grad_norm": 0.19577565789222717, + "learning_rate": 0.0003412066497458158, + "loss": 1.8143, + "step": 4325 + }, + { + "epoch": 0.4224609375, + "grad_norm": 0.18919500708580017, + "learning_rate": 0.0003411386822278779, + "loss": 1.8241, + "step": 4326 + }, + { + "epoch": 0.42255859375, + "grad_norm": 0.248185396194458, + "learning_rate": 0.0003410707081032717, + "loss": 1.8358, + "step": 4327 + }, + { + "epoch": 0.42265625, + "grad_norm": 0.17570361495018005, + "learning_rate": 0.00034100272737878706, + "loss": 1.8312, + "step": 4328 + }, + { + "epoch": 0.42275390625, + "grad_norm": 0.24675165116786957, + "learning_rate": 0.00034093474006121477, + "loss": 1.8289, + "step": 4329 + }, + { + "epoch": 0.4228515625, + "grad_norm": 0.1773347556591034, + "learning_rate": 0.000340866746157346, + "loss": 1.8583, + "step": 4330 + }, + { + "epoch": 0.42294921875, + "grad_norm": 0.2368694543838501, + "learning_rate": 0.00034079874567397283, + "loss": 1.8379, + "step": 4331 + }, + { + "epoch": 0.423046875, + "grad_norm": 0.21391184628009796, + "learning_rate": 0.00034073073861788785, + "loss": 1.8469, + "step": 4332 + }, + { + "epoch": 0.42314453125, + "grad_norm": 0.24347218871116638, + "learning_rate": 0.00034066272499588444, + "loss": 1.8379, + "step": 4333 + }, + { + "epoch": 0.4232421875, + "grad_norm": 0.20501993596553802, + "learning_rate": 0.0003405947048147566, + "loss": 1.8698, + "step": 4334 + }, + { + "epoch": 0.42333984375, + "grad_norm": 0.23922660946846008, + "learning_rate": 0.00034052667808129864, + "loss": 1.8463, + "step": 4335 + }, + { + "epoch": 0.4234375, + "grad_norm": 0.22898203134536743, + "learning_rate": 0.00034045864480230625, + "loss": 1.838, + "step": 4336 + }, + { + "epoch": 0.42353515625, + "grad_norm": 0.2185804545879364, + "learning_rate": 0.0003403906049845751, + "loss": 1.8265, + "step": 4337 + }, + { + "epoch": 0.4236328125, + "grad_norm": 0.2115095853805542, + "learning_rate": 0.00034032255863490184, + "loss": 1.8322, + "step": 4338 + }, + { + "epoch": 0.42373046875, + "grad_norm": 0.1770572066307068, + "learning_rate": 0.00034025450576008373, + "loss": 1.8171, + "step": 4339 + }, + { + "epoch": 0.423828125, + "grad_norm": 0.2317141890525818, + "learning_rate": 0.0003401864463669185, + "loss": 1.8298, + "step": 4340 + }, + { + "epoch": 0.42392578125, + "grad_norm": 0.20465056598186493, + "learning_rate": 0.00034011838046220486, + "loss": 1.8688, + "step": 4341 + }, + { + "epoch": 0.4240234375, + "grad_norm": 0.1989961564540863, + "learning_rate": 0.0003400503080527418, + "loss": 1.8509, + "step": 4342 + }, + { + "epoch": 0.42412109375, + "grad_norm": 0.2024098038673401, + "learning_rate": 0.00033998222914532935, + "loss": 1.8697, + "step": 4343 + }, + { + "epoch": 0.42421875, + "grad_norm": 0.19737909734249115, + "learning_rate": 0.0003399141437467678, + "loss": 1.8403, + "step": 4344 + }, + { + "epoch": 0.42431640625, + "grad_norm": 0.19448748230934143, + "learning_rate": 0.0003398460518638585, + "loss": 1.8132, + "step": 4345 + }, + { + "epoch": 0.4244140625, + "grad_norm": 0.231892392039299, + "learning_rate": 0.00033977795350340306, + "loss": 1.8447, + "step": 4346 + }, + { + "epoch": 0.42451171875, + "grad_norm": 0.24322080612182617, + "learning_rate": 0.0003397098486722039, + "loss": 1.8636, + "step": 4347 + }, + { + "epoch": 0.424609375, + "grad_norm": 0.20982776582241058, + "learning_rate": 0.0003396417373770642, + "loss": 1.8177, + "step": 4348 + }, + { + "epoch": 0.42470703125, + "grad_norm": 0.17859229445457458, + "learning_rate": 0.0003395736196247875, + "loss": 1.8398, + "step": 4349 + }, + { + "epoch": 0.4248046875, + "grad_norm": 0.17166008055210114, + "learning_rate": 0.0003395054954221784, + "loss": 1.8346, + "step": 4350 + }, + { + "epoch": 0.42490234375, + "grad_norm": 0.192952960729599, + "learning_rate": 0.0003394373647760417, + "loss": 1.856, + "step": 4351 + }, + { + "epoch": 0.425, + "grad_norm": 0.18562056124210358, + "learning_rate": 0.00033936922769318317, + "loss": 1.8328, + "step": 4352 + }, + { + "epoch": 0.42509765625, + "grad_norm": 0.18785525858402252, + "learning_rate": 0.00033930108418040903, + "loss": 1.8347, + "step": 4353 + }, + { + "epoch": 0.4251953125, + "grad_norm": 0.20413319766521454, + "learning_rate": 0.0003392329342445262, + "loss": 1.8773, + "step": 4354 + }, + { + "epoch": 0.42529296875, + "grad_norm": 0.227433443069458, + "learning_rate": 0.0003391647778923424, + "loss": 1.8488, + "step": 4355 + }, + { + "epoch": 0.425390625, + "grad_norm": 0.1805514395236969, + "learning_rate": 0.0003390966151306656, + "loss": 1.8409, + "step": 4356 + }, + { + "epoch": 0.42548828125, + "grad_norm": 0.21339501440525055, + "learning_rate": 0.0003390284459663049, + "loss": 1.8263, + "step": 4357 + }, + { + "epoch": 0.4255859375, + "grad_norm": 0.21802109479904175, + "learning_rate": 0.0003389602704060696, + "loss": 1.8303, + "step": 4358 + }, + { + "epoch": 0.42568359375, + "grad_norm": 0.1962195634841919, + "learning_rate": 0.00033889208845676994, + "loss": 1.8547, + "step": 4359 + }, + { + "epoch": 0.42578125, + "grad_norm": 0.20132724940776825, + "learning_rate": 0.00033882390012521675, + "loss": 1.8268, + "step": 4360 + }, + { + "epoch": 0.42587890625, + "grad_norm": 0.2011682391166687, + "learning_rate": 0.0003387557054182214, + "loss": 1.8384, + "step": 4361 + }, + { + "epoch": 0.4259765625, + "grad_norm": 0.1959267556667328, + "learning_rate": 0.0003386875043425958, + "loss": 1.8443, + "step": 4362 + }, + { + "epoch": 0.42607421875, + "grad_norm": 0.20704340934753418, + "learning_rate": 0.0003386192969051527, + "loss": 1.8025, + "step": 4363 + }, + { + "epoch": 0.426171875, + "grad_norm": 0.22771066427230835, + "learning_rate": 0.0003385510831127056, + "loss": 1.8652, + "step": 4364 + }, + { + "epoch": 0.42626953125, + "grad_norm": 0.20367157459259033, + "learning_rate": 0.00033848286297206813, + "loss": 1.8589, + "step": 4365 + }, + { + "epoch": 0.4263671875, + "grad_norm": 0.21674802899360657, + "learning_rate": 0.00033841463649005515, + "loss": 1.8249, + "step": 4366 + }, + { + "epoch": 0.42646484375, + "grad_norm": 0.23962077498435974, + "learning_rate": 0.0003383464036734818, + "loss": 1.8522, + "step": 4367 + }, + { + "epoch": 0.4265625, + "grad_norm": 0.2480829656124115, + "learning_rate": 0.00033827816452916396, + "loss": 1.8578, + "step": 4368 + }, + { + "epoch": 0.42666015625, + "grad_norm": 0.23992200195789337, + "learning_rate": 0.00033820991906391814, + "loss": 1.847, + "step": 4369 + }, + { + "epoch": 0.4267578125, + "grad_norm": 0.16551890969276428, + "learning_rate": 0.00033814166728456113, + "loss": 1.8748, + "step": 4370 + }, + { + "epoch": 0.42685546875, + "grad_norm": 0.2550032138824463, + "learning_rate": 0.0003380734091979111, + "loss": 1.8253, + "step": 4371 + }, + { + "epoch": 0.426953125, + "grad_norm": 0.2584201395511627, + "learning_rate": 0.0003380051448107863, + "loss": 1.8391, + "step": 4372 + }, + { + "epoch": 0.42705078125, + "grad_norm": 0.17704187333583832, + "learning_rate": 0.0003379368741300057, + "loss": 1.8121, + "step": 4373 + }, + { + "epoch": 0.4271484375, + "grad_norm": 0.2456081509590149, + "learning_rate": 0.0003378685971623889, + "loss": 1.8527, + "step": 4374 + }, + { + "epoch": 0.42724609375, + "grad_norm": 0.25017473101615906, + "learning_rate": 0.00033780031391475625, + "loss": 1.8178, + "step": 4375 + }, + { + "epoch": 0.42734375, + "grad_norm": 0.18413616716861725, + "learning_rate": 0.0003377320243939286, + "loss": 1.8327, + "step": 4376 + }, + { + "epoch": 0.42744140625, + "grad_norm": 0.28050103783607483, + "learning_rate": 0.0003376637286067274, + "loss": 1.8438, + "step": 4377 + }, + { + "epoch": 0.4275390625, + "grad_norm": 0.23926864564418793, + "learning_rate": 0.00033759542655997495, + "loss": 1.7982, + "step": 4378 + }, + { + "epoch": 0.42763671875, + "grad_norm": 0.24252592027187347, + "learning_rate": 0.0003375271182604939, + "loss": 1.8424, + "step": 4379 + }, + { + "epoch": 0.427734375, + "grad_norm": 0.2253219336271286, + "learning_rate": 0.00033745880371510766, + "loss": 1.8144, + "step": 4380 + }, + { + "epoch": 0.42783203125, + "grad_norm": 0.2022935003042221, + "learning_rate": 0.0003373904829306402, + "loss": 1.8445, + "step": 4381 + }, + { + "epoch": 0.4279296875, + "grad_norm": 0.23967212438583374, + "learning_rate": 0.0003373221559139164, + "loss": 1.8364, + "step": 4382 + }, + { + "epoch": 0.42802734375, + "grad_norm": 0.2502332925796509, + "learning_rate": 0.00033725382267176126, + "loss": 1.8337, + "step": 4383 + }, + { + "epoch": 0.428125, + "grad_norm": 0.20870167016983032, + "learning_rate": 0.0003371854832110007, + "loss": 1.7944, + "step": 4384 + }, + { + "epoch": 0.42822265625, + "grad_norm": 0.23509882390499115, + "learning_rate": 0.0003371171375384614, + "loss": 1.783, + "step": 4385 + }, + { + "epoch": 0.4283203125, + "grad_norm": 0.21745553612709045, + "learning_rate": 0.00033704878566097026, + "loss": 1.8586, + "step": 4386 + }, + { + "epoch": 0.42841796875, + "grad_norm": 0.23707370460033417, + "learning_rate": 0.0003369804275853551, + "loss": 1.8661, + "step": 4387 + }, + { + "epoch": 0.428515625, + "grad_norm": 0.2220204472541809, + "learning_rate": 0.00033691206331844443, + "loss": 1.8088, + "step": 4388 + }, + { + "epoch": 0.42861328125, + "grad_norm": 0.21647056937217712, + "learning_rate": 0.00033684369286706713, + "loss": 1.8383, + "step": 4389 + }, + { + "epoch": 0.4287109375, + "grad_norm": 0.21368125081062317, + "learning_rate": 0.00033677531623805277, + "loss": 1.8219, + "step": 4390 + }, + { + "epoch": 0.42880859375, + "grad_norm": 0.28654998540878296, + "learning_rate": 0.00033670693343823153, + "loss": 1.8687, + "step": 4391 + }, + { + "epoch": 0.42890625, + "grad_norm": 0.19630244374275208, + "learning_rate": 0.00033663854447443445, + "loss": 1.8421, + "step": 4392 + }, + { + "epoch": 0.42900390625, + "grad_norm": 0.2457696795463562, + "learning_rate": 0.00033657014935349275, + "loss": 1.8398, + "step": 4393 + }, + { + "epoch": 0.4291015625, + "grad_norm": 0.2542276680469513, + "learning_rate": 0.0003365017480822385, + "loss": 1.8567, + "step": 4394 + }, + { + "epoch": 0.42919921875, + "grad_norm": 0.22448401153087616, + "learning_rate": 0.0003364333406675046, + "loss": 1.8188, + "step": 4395 + }, + { + "epoch": 0.429296875, + "grad_norm": 0.2244417518377304, + "learning_rate": 0.0003363649271161243, + "loss": 1.798, + "step": 4396 + }, + { + "epoch": 0.42939453125, + "grad_norm": 0.24035105109214783, + "learning_rate": 0.0003362965074349313, + "loss": 1.8418, + "step": 4397 + }, + { + "epoch": 0.4294921875, + "grad_norm": 0.2114304155111313, + "learning_rate": 0.0003362280816307602, + "loss": 1.8185, + "step": 4398 + }, + { + "epoch": 0.42958984375, + "grad_norm": 0.22470279037952423, + "learning_rate": 0.0003361596497104463, + "loss": 1.8282, + "step": 4399 + }, + { + "epoch": 0.4296875, + "grad_norm": 0.22919301688671112, + "learning_rate": 0.0003360912116808251, + "loss": 1.8662, + "step": 4400 + }, + { + "epoch": 0.42978515625, + "grad_norm": 0.24255748093128204, + "learning_rate": 0.0003360227675487332, + "loss": 1.8386, + "step": 4401 + }, + { + "epoch": 0.4298828125, + "grad_norm": 0.2792574167251587, + "learning_rate": 0.0003359543173210073, + "loss": 1.8308, + "step": 4402 + }, + { + "epoch": 0.42998046875, + "grad_norm": 0.22785964608192444, + "learning_rate": 0.0003358858610044852, + "loss": 1.8419, + "step": 4403 + }, + { + "epoch": 0.430078125, + "grad_norm": 0.23355011641979218, + "learning_rate": 0.00033581739860600495, + "loss": 1.8332, + "step": 4404 + }, + { + "epoch": 0.43017578125, + "grad_norm": 0.2188209593296051, + "learning_rate": 0.00033574893013240537, + "loss": 1.8468, + "step": 4405 + }, + { + "epoch": 0.4302734375, + "grad_norm": 0.21975260972976685, + "learning_rate": 0.00033568045559052596, + "loss": 1.8454, + "step": 4406 + }, + { + "epoch": 0.43037109375, + "grad_norm": 0.22873692214488983, + "learning_rate": 0.00033561197498720646, + "loss": 1.8758, + "step": 4407 + }, + { + "epoch": 0.43046875, + "grad_norm": 0.25508803129196167, + "learning_rate": 0.0003355434883292877, + "loss": 1.882, + "step": 4408 + }, + { + "epoch": 0.43056640625, + "grad_norm": 0.22015030682086945, + "learning_rate": 0.00033547499562361084, + "loss": 1.835, + "step": 4409 + }, + { + "epoch": 0.4306640625, + "grad_norm": 0.2391720563173294, + "learning_rate": 0.0003354064968770177, + "loss": 1.8394, + "step": 4410 + }, + { + "epoch": 0.43076171875, + "grad_norm": 0.23599278926849365, + "learning_rate": 0.0003353379920963506, + "loss": 1.8123, + "step": 4411 + }, + { + "epoch": 0.430859375, + "grad_norm": 0.2925673723220825, + "learning_rate": 0.00033526948128845263, + "loss": 1.859, + "step": 4412 + }, + { + "epoch": 0.43095703125, + "grad_norm": 0.2177896946668625, + "learning_rate": 0.0003352009644601675, + "loss": 1.812, + "step": 4413 + }, + { + "epoch": 0.4310546875, + "grad_norm": 0.18912582099437714, + "learning_rate": 0.0003351324416183393, + "loss": 1.8409, + "step": 4414 + }, + { + "epoch": 0.43115234375, + "grad_norm": 0.197793111205101, + "learning_rate": 0.00033506391276981294, + "loss": 1.9291, + "step": 4415 + }, + { + "epoch": 0.43125, + "grad_norm": 0.21789690852165222, + "learning_rate": 0.0003349953779214338, + "loss": 1.8663, + "step": 4416 + }, + { + "epoch": 0.43134765625, + "grad_norm": 0.19930073618888855, + "learning_rate": 0.00033492683708004795, + "loss": 1.872, + "step": 4417 + }, + { + "epoch": 0.4314453125, + "grad_norm": 0.19942253828048706, + "learning_rate": 0.00033485829025250194, + "loss": 1.8517, + "step": 4418 + }, + { + "epoch": 0.43154296875, + "grad_norm": 0.21190962195396423, + "learning_rate": 0.000334789737445643, + "loss": 1.8754, + "step": 4419 + }, + { + "epoch": 0.431640625, + "grad_norm": 0.17792046070098877, + "learning_rate": 0.00033472117866631905, + "loss": 1.8699, + "step": 4420 + }, + { + "epoch": 0.43173828125, + "grad_norm": 0.2024158537387848, + "learning_rate": 0.00033465261392137837, + "loss": 1.8027, + "step": 4421 + }, + { + "epoch": 0.4318359375, + "grad_norm": 0.16588537395000458, + "learning_rate": 0.0003345840432176701, + "loss": 1.8071, + "step": 4422 + }, + { + "epoch": 0.43193359375, + "grad_norm": 0.1971169114112854, + "learning_rate": 0.0003345154665620437, + "loss": 1.7873, + "step": 4423 + }, + { + "epoch": 0.43203125, + "grad_norm": 0.1795888990163803, + "learning_rate": 0.0003344468839613495, + "loss": 1.8088, + "step": 4424 + }, + { + "epoch": 0.43212890625, + "grad_norm": 0.1773122400045395, + "learning_rate": 0.0003343782954224382, + "loss": 1.8417, + "step": 4425 + }, + { + "epoch": 0.4322265625, + "grad_norm": 0.1802736520767212, + "learning_rate": 0.0003343097009521613, + "loss": 1.8258, + "step": 4426 + }, + { + "epoch": 0.43232421875, + "grad_norm": 0.19189099967479706, + "learning_rate": 0.0003342411005573707, + "loss": 1.849, + "step": 4427 + }, + { + "epoch": 0.432421875, + "grad_norm": 0.21457602083683014, + "learning_rate": 0.0003341724942449189, + "loss": 1.8498, + "step": 4428 + }, + { + "epoch": 0.43251953125, + "grad_norm": 0.20727668702602386, + "learning_rate": 0.0003341038820216592, + "loss": 1.7926, + "step": 4429 + }, + { + "epoch": 0.4326171875, + "grad_norm": 0.26479217410087585, + "learning_rate": 0.00033403526389444535, + "loss": 1.8665, + "step": 4430 + }, + { + "epoch": 0.43271484375, + "grad_norm": 0.1955796629190445, + "learning_rate": 0.0003339666398701316, + "loss": 1.8521, + "step": 4431 + }, + { + "epoch": 0.4328125, + "grad_norm": 0.22441504895687103, + "learning_rate": 0.0003338980099555729, + "loss": 1.8614, + "step": 4432 + }, + { + "epoch": 0.43291015625, + "grad_norm": 0.20976248383522034, + "learning_rate": 0.0003338293741576248, + "loss": 1.8114, + "step": 4433 + }, + { + "epoch": 0.4330078125, + "grad_norm": 0.20051336288452148, + "learning_rate": 0.0003337607324831434, + "loss": 1.848, + "step": 4434 + }, + { + "epoch": 0.43310546875, + "grad_norm": 0.20708587765693665, + "learning_rate": 0.00033369208493898544, + "loss": 1.8375, + "step": 4435 + }, + { + "epoch": 0.433203125, + "grad_norm": 0.23173046112060547, + "learning_rate": 0.0003336234315320081, + "loss": 1.8573, + "step": 4436 + }, + { + "epoch": 0.43330078125, + "grad_norm": 0.2623552083969116, + "learning_rate": 0.00033355477226906934, + "loss": 1.8618, + "step": 4437 + }, + { + "epoch": 0.4333984375, + "grad_norm": 0.2273312360048294, + "learning_rate": 0.0003334861071570276, + "loss": 1.8438, + "step": 4438 + }, + { + "epoch": 0.43349609375, + "grad_norm": 0.16699610650539398, + "learning_rate": 0.0003334174362027419, + "loss": 1.7673, + "step": 4439 + }, + { + "epoch": 0.43359375, + "grad_norm": 0.20863619446754456, + "learning_rate": 0.0003333487594130718, + "loss": 1.8566, + "step": 4440 + }, + { + "epoch": 0.43369140625, + "grad_norm": 0.2478940486907959, + "learning_rate": 0.0003332800767948776, + "loss": 1.815, + "step": 4441 + }, + { + "epoch": 0.4337890625, + "grad_norm": 0.24342428147792816, + "learning_rate": 0.00033321138835502005, + "loss": 1.844, + "step": 4442 + }, + { + "epoch": 0.43388671875, + "grad_norm": 0.2302752137184143, + "learning_rate": 0.0003331426941003605, + "loss": 1.8641, + "step": 4443 + }, + { + "epoch": 0.433984375, + "grad_norm": 0.16929592192173004, + "learning_rate": 0.0003330739940377608, + "loss": 1.861, + "step": 4444 + }, + { + "epoch": 0.43408203125, + "grad_norm": 0.22187499701976776, + "learning_rate": 0.0003330052881740838, + "loss": 1.8441, + "step": 4445 + }, + { + "epoch": 0.4341796875, + "grad_norm": 0.19175705313682556, + "learning_rate": 0.00033293657651619225, + "loss": 1.8116, + "step": 4446 + }, + { + "epoch": 0.43427734375, + "grad_norm": 0.1969825178384781, + "learning_rate": 0.00033286785907094994, + "loss": 1.8518, + "step": 4447 + }, + { + "epoch": 0.434375, + "grad_norm": 0.20496003329753876, + "learning_rate": 0.0003327991358452213, + "loss": 1.8396, + "step": 4448 + }, + { + "epoch": 0.43447265625, + "grad_norm": 0.2111838310956955, + "learning_rate": 0.000332730406845871, + "loss": 1.8398, + "step": 4449 + }, + { + "epoch": 0.4345703125, + "grad_norm": 0.18950366973876953, + "learning_rate": 0.00033266167207976445, + "loss": 1.8106, + "step": 4450 + }, + { + "epoch": 0.43466796875, + "grad_norm": 0.20442768931388855, + "learning_rate": 0.00033259293155376776, + "loss": 1.8547, + "step": 4451 + }, + { + "epoch": 0.434765625, + "grad_norm": 0.2146398425102234, + "learning_rate": 0.0003325241852747474, + "loss": 1.8698, + "step": 4452 + }, + { + "epoch": 0.43486328125, + "grad_norm": 0.2124415785074234, + "learning_rate": 0.0003324554332495705, + "loss": 1.7994, + "step": 4453 + }, + { + "epoch": 0.4349609375, + "grad_norm": 0.21013392508029938, + "learning_rate": 0.0003323866754851049, + "loss": 1.8857, + "step": 4454 + }, + { + "epoch": 0.43505859375, + "grad_norm": 0.2193000763654709, + "learning_rate": 0.0003323179119882189, + "loss": 1.8585, + "step": 4455 + }, + { + "epoch": 0.43515625, + "grad_norm": 0.19693540036678314, + "learning_rate": 0.0003322491427657811, + "loss": 1.823, + "step": 4456 + }, + { + "epoch": 0.43525390625, + "grad_norm": 0.19659705460071564, + "learning_rate": 0.00033218036782466115, + "loss": 1.8586, + "step": 4457 + }, + { + "epoch": 0.4353515625, + "grad_norm": 0.22691860795021057, + "learning_rate": 0.00033211158717172905, + "loss": 1.7934, + "step": 4458 + }, + { + "epoch": 0.43544921875, + "grad_norm": 0.2183864265680313, + "learning_rate": 0.00033204280081385544, + "loss": 1.8764, + "step": 4459 + }, + { + "epoch": 0.435546875, + "grad_norm": 0.2263919562101364, + "learning_rate": 0.00033197400875791135, + "loss": 1.8278, + "step": 4460 + }, + { + "epoch": 0.43564453125, + "grad_norm": 0.2252654880285263, + "learning_rate": 0.0003319052110107685, + "loss": 1.8433, + "step": 4461 + }, + { + "epoch": 0.4357421875, + "grad_norm": 0.2387869954109192, + "learning_rate": 0.00033183640757929926, + "loss": 1.8106, + "step": 4462 + }, + { + "epoch": 0.43583984375, + "grad_norm": 0.2297157645225525, + "learning_rate": 0.00033176759847037635, + "loss": 1.8814, + "step": 4463 + }, + { + "epoch": 0.4359375, + "grad_norm": 0.2321292757987976, + "learning_rate": 0.0003316987836908733, + "loss": 1.8298, + "step": 4464 + }, + { + "epoch": 0.43603515625, + "grad_norm": 0.2759620249271393, + "learning_rate": 0.00033162996324766404, + "loss": 1.8671, + "step": 4465 + }, + { + "epoch": 0.4361328125, + "grad_norm": 0.20262287557125092, + "learning_rate": 0.0003315611371476232, + "loss": 1.802, + "step": 4466 + }, + { + "epoch": 0.43623046875, + "grad_norm": 0.23232300579547882, + "learning_rate": 0.00033149230539762577, + "loss": 1.8417, + "step": 4467 + }, + { + "epoch": 0.436328125, + "grad_norm": 0.21800456941127777, + "learning_rate": 0.0003314234680045475, + "loss": 1.8644, + "step": 4468 + }, + { + "epoch": 0.43642578125, + "grad_norm": 0.19706159830093384, + "learning_rate": 0.00033135462497526475, + "loss": 1.8082, + "step": 4469 + }, + { + "epoch": 0.4365234375, + "grad_norm": 0.2481984794139862, + "learning_rate": 0.0003312857763166542, + "loss": 1.8404, + "step": 4470 + }, + { + "epoch": 0.43662109375, + "grad_norm": 0.22650815546512604, + "learning_rate": 0.00033121692203559323, + "loss": 1.8619, + "step": 4471 + }, + { + "epoch": 0.43671875, + "grad_norm": 0.2094004899263382, + "learning_rate": 0.0003311480621389597, + "loss": 1.8826, + "step": 4472 + }, + { + "epoch": 0.43681640625, + "grad_norm": 0.24898546934127808, + "learning_rate": 0.0003310791966336323, + "loss": 1.8522, + "step": 4473 + }, + { + "epoch": 0.4369140625, + "grad_norm": 0.20734456181526184, + "learning_rate": 0.0003310103255264899, + "loss": 1.7986, + "step": 4474 + }, + { + "epoch": 0.43701171875, + "grad_norm": 0.23696695268154144, + "learning_rate": 0.0003309414488244122, + "loss": 1.8541, + "step": 4475 + }, + { + "epoch": 0.437109375, + "grad_norm": 0.24068109691143036, + "learning_rate": 0.00033087256653427946, + "loss": 1.8579, + "step": 4476 + }, + { + "epoch": 0.43720703125, + "grad_norm": 0.19585324823856354, + "learning_rate": 0.00033080367866297213, + "loss": 1.8765, + "step": 4477 + }, + { + "epoch": 0.4373046875, + "grad_norm": 0.2366010844707489, + "learning_rate": 0.00033073478521737176, + "loss": 1.8456, + "step": 4478 + }, + { + "epoch": 0.43740234375, + "grad_norm": 0.18371711671352386, + "learning_rate": 0.00033066588620436017, + "loss": 1.817, + "step": 4479 + }, + { + "epoch": 0.4375, + "grad_norm": 0.24547503888607025, + "learning_rate": 0.0003305969816308197, + "loss": 1.8545, + "step": 4480 + }, + { + "epoch": 0.43759765625, + "grad_norm": 0.22530093789100647, + "learning_rate": 0.00033052807150363316, + "loss": 1.8209, + "step": 4481 + }, + { + "epoch": 0.4376953125, + "grad_norm": 0.22203901410102844, + "learning_rate": 0.0003304591558296844, + "loss": 1.83, + "step": 4482 + }, + { + "epoch": 0.43779296875, + "grad_norm": 0.2331925481557846, + "learning_rate": 0.0003303902346158572, + "loss": 1.8952, + "step": 4483 + }, + { + "epoch": 0.437890625, + "grad_norm": 0.26899224519729614, + "learning_rate": 0.0003303213078690363, + "loss": 1.8432, + "step": 4484 + }, + { + "epoch": 0.43798828125, + "grad_norm": 0.8147704005241394, + "learning_rate": 0.00033025237559610685, + "loss": 1.9056, + "step": 4485 + }, + { + "epoch": 0.4380859375, + "grad_norm": 0.26135900616645813, + "learning_rate": 0.0003301834378039545, + "loss": 1.8493, + "step": 4486 + }, + { + "epoch": 0.43818359375, + "grad_norm": 0.2273353487253189, + "learning_rate": 0.00033011449449946574, + "loss": 1.8561, + "step": 4487 + }, + { + "epoch": 0.43828125, + "grad_norm": 0.2513292133808136, + "learning_rate": 0.00033004554568952704, + "loss": 1.851, + "step": 4488 + }, + { + "epoch": 0.43837890625, + "grad_norm": 0.20417039096355438, + "learning_rate": 0.0003299765913810261, + "loss": 1.8483, + "step": 4489 + }, + { + "epoch": 0.4384765625, + "grad_norm": 0.24769926071166992, + "learning_rate": 0.0003299076315808508, + "loss": 1.8546, + "step": 4490 + }, + { + "epoch": 0.43857421875, + "grad_norm": 0.21822988986968994, + "learning_rate": 0.00032983866629588944, + "loss": 1.8686, + "step": 4491 + }, + { + "epoch": 0.438671875, + "grad_norm": 0.25055810809135437, + "learning_rate": 0.0003297696955330313, + "loss": 1.8126, + "step": 4492 + }, + { + "epoch": 0.43876953125, + "grad_norm": 0.1908411830663681, + "learning_rate": 0.00032970071929916563, + "loss": 1.8546, + "step": 4493 + }, + { + "epoch": 0.4388671875, + "grad_norm": 0.19980601966381073, + "learning_rate": 0.00032963173760118276, + "loss": 1.7861, + "step": 4494 + }, + { + "epoch": 0.43896484375, + "grad_norm": 0.18650661408901215, + "learning_rate": 0.0003295627504459732, + "loss": 1.829, + "step": 4495 + }, + { + "epoch": 0.4390625, + "grad_norm": 0.19553345441818237, + "learning_rate": 0.0003294937578404284, + "loss": 1.8121, + "step": 4496 + }, + { + "epoch": 0.43916015625, + "grad_norm": 0.20082074403762817, + "learning_rate": 0.00032942475979143987, + "loss": 1.8404, + "step": 4497 + }, + { + "epoch": 0.4392578125, + "grad_norm": 0.20696525275707245, + "learning_rate": 0.00032935575630590005, + "loss": 1.8435, + "step": 4498 + }, + { + "epoch": 0.43935546875, + "grad_norm": 0.2295655757188797, + "learning_rate": 0.00032928674739070177, + "loss": 1.8257, + "step": 4499 + }, + { + "epoch": 0.439453125, + "grad_norm": 0.22089581191539764, + "learning_rate": 0.0003292177330527382, + "loss": 1.8122, + "step": 4500 + }, + { + "epoch": 0.43955078125, + "grad_norm": 0.21159827709197998, + "learning_rate": 0.0003291487132989036, + "loss": 1.8339, + "step": 4501 + }, + { + "epoch": 0.4396484375, + "grad_norm": 0.2064434140920639, + "learning_rate": 0.00032907968813609213, + "loss": 1.7816, + "step": 4502 + }, + { + "epoch": 0.43974609375, + "grad_norm": 0.2011205554008484, + "learning_rate": 0.000329010657571199, + "loss": 1.8245, + "step": 4503 + }, + { + "epoch": 0.43984375, + "grad_norm": 0.23691421747207642, + "learning_rate": 0.00032894162161111964, + "loss": 1.9096, + "step": 4504 + }, + { + "epoch": 0.43994140625, + "grad_norm": 0.19454586505889893, + "learning_rate": 0.00032887258026275014, + "loss": 1.8582, + "step": 4505 + }, + { + "epoch": 0.4400390625, + "grad_norm": 0.18718378245830536, + "learning_rate": 0.0003288035335329871, + "loss": 1.8708, + "step": 4506 + }, + { + "epoch": 0.44013671875, + "grad_norm": 0.19976142048835754, + "learning_rate": 0.00032873448142872784, + "loss": 1.8312, + "step": 4507 + }, + { + "epoch": 0.440234375, + "grad_norm": 0.1672353744506836, + "learning_rate": 0.0003286654239568699, + "loss": 1.8541, + "step": 4508 + }, + { + "epoch": 0.44033203125, + "grad_norm": 0.19212840497493744, + "learning_rate": 0.00032859636112431145, + "loss": 1.8372, + "step": 4509 + }, + { + "epoch": 0.4404296875, + "grad_norm": 0.2028074860572815, + "learning_rate": 0.0003285272929379514, + "loss": 1.8201, + "step": 4510 + }, + { + "epoch": 0.44052734375, + "grad_norm": 0.2094787210226059, + "learning_rate": 0.00032845821940468903, + "loss": 1.7874, + "step": 4511 + }, + { + "epoch": 0.440625, + "grad_norm": 0.19495780766010284, + "learning_rate": 0.00032838914053142407, + "loss": 1.8786, + "step": 4512 + }, + { + "epoch": 0.44072265625, + "grad_norm": 0.21458013355731964, + "learning_rate": 0.00032832005632505703, + "loss": 1.8543, + "step": 4513 + }, + { + "epoch": 0.4408203125, + "grad_norm": 0.19527952373027802, + "learning_rate": 0.00032825096679248866, + "loss": 1.8582, + "step": 4514 + }, + { + "epoch": 0.44091796875, + "grad_norm": 0.2177453637123108, + "learning_rate": 0.00032818187194062054, + "loss": 1.8508, + "step": 4515 + }, + { + "epoch": 0.441015625, + "grad_norm": 0.19066031277179718, + "learning_rate": 0.00032811277177635445, + "loss": 1.8406, + "step": 4516 + }, + { + "epoch": 0.44111328125, + "grad_norm": 0.2284795641899109, + "learning_rate": 0.0003280436663065931, + "loss": 1.8698, + "step": 4517 + }, + { + "epoch": 0.4412109375, + "grad_norm": 0.1739828735589981, + "learning_rate": 0.0003279745555382394, + "loss": 1.8475, + "step": 4518 + }, + { + "epoch": 0.44130859375, + "grad_norm": 0.20771338045597076, + "learning_rate": 0.00032790543947819686, + "loss": 1.8242, + "step": 4519 + }, + { + "epoch": 0.44140625, + "grad_norm": 0.174216166138649, + "learning_rate": 0.00032783631813336966, + "loss": 1.8561, + "step": 4520 + }, + { + "epoch": 0.44150390625, + "grad_norm": 0.21372954547405243, + "learning_rate": 0.0003277671915106624, + "loss": 1.8698, + "step": 4521 + }, + { + "epoch": 0.4416015625, + "grad_norm": 0.23978406190872192, + "learning_rate": 0.0003276980596169801, + "loss": 1.8271, + "step": 4522 + }, + { + "epoch": 0.44169921875, + "grad_norm": 0.24592150747776031, + "learning_rate": 0.00032762892245922847, + "loss": 1.819, + "step": 4523 + }, + { + "epoch": 0.441796875, + "grad_norm": 0.23378491401672363, + "learning_rate": 0.0003275597800443139, + "loss": 1.8596, + "step": 4524 + }, + { + "epoch": 0.44189453125, + "grad_norm": 0.173017218708992, + "learning_rate": 0.00032749063237914286, + "loss": 1.8465, + "step": 4525 + }, + { + "epoch": 0.4419921875, + "grad_norm": 0.2911878228187561, + "learning_rate": 0.00032742147947062265, + "loss": 1.8099, + "step": 4526 + }, + { + "epoch": 0.44208984375, + "grad_norm": 0.23090296983718872, + "learning_rate": 0.00032735232132566106, + "loss": 1.8158, + "step": 4527 + }, + { + "epoch": 0.4421875, + "grad_norm": 0.20182637870311737, + "learning_rate": 0.00032728315795116635, + "loss": 1.8516, + "step": 4528 + }, + { + "epoch": 0.44228515625, + "grad_norm": 0.22502705454826355, + "learning_rate": 0.00032721398935404744, + "loss": 1.8343, + "step": 4529 + }, + { + "epoch": 0.4423828125, + "grad_norm": 0.2155645191669464, + "learning_rate": 0.00032714481554121347, + "loss": 1.8259, + "step": 4530 + }, + { + "epoch": 0.44248046875, + "grad_norm": 0.20365890860557556, + "learning_rate": 0.00032707563651957443, + "loss": 1.8103, + "step": 4531 + }, + { + "epoch": 0.442578125, + "grad_norm": 0.20811273157596588, + "learning_rate": 0.00032700645229604065, + "loss": 1.8743, + "step": 4532 + }, + { + "epoch": 0.44267578125, + "grad_norm": 0.2342660278081894, + "learning_rate": 0.00032693726287752294, + "loss": 1.8403, + "step": 4533 + }, + { + "epoch": 0.4427734375, + "grad_norm": 0.15678024291992188, + "learning_rate": 0.000326868068270933, + "loss": 1.8119, + "step": 4534 + }, + { + "epoch": 0.44287109375, + "grad_norm": 0.257323682308197, + "learning_rate": 0.0003267988684831824, + "loss": 1.8259, + "step": 4535 + }, + { + "epoch": 0.44296875, + "grad_norm": 0.21096593141555786, + "learning_rate": 0.00032672966352118374, + "loss": 1.8489, + "step": 4536 + }, + { + "epoch": 0.44306640625, + "grad_norm": 0.2324661761522293, + "learning_rate": 0.00032666045339184994, + "loss": 1.8353, + "step": 4537 + }, + { + "epoch": 0.4431640625, + "grad_norm": 0.20259353518486023, + "learning_rate": 0.0003265912381020946, + "loss": 1.8492, + "step": 4538 + }, + { + "epoch": 0.44326171875, + "grad_norm": 0.2664618492126465, + "learning_rate": 0.0003265220176588315, + "loss": 1.8471, + "step": 4539 + }, + { + "epoch": 0.443359375, + "grad_norm": 0.22948837280273438, + "learning_rate": 0.00032645279206897543, + "loss": 1.8128, + "step": 4540 + }, + { + "epoch": 0.44345703125, + "grad_norm": 0.23944346606731415, + "learning_rate": 0.00032638356133944126, + "loss": 1.817, + "step": 4541 + }, + { + "epoch": 0.4435546875, + "grad_norm": 0.2528168261051178, + "learning_rate": 0.00032631432547714444, + "loss": 1.818, + "step": 4542 + }, + { + "epoch": 0.44365234375, + "grad_norm": 0.22383758425712585, + "learning_rate": 0.00032624508448900115, + "loss": 1.8603, + "step": 4543 + }, + { + "epoch": 0.44375, + "grad_norm": 0.2284606248140335, + "learning_rate": 0.00032617583838192795, + "loss": 1.8358, + "step": 4544 + }, + { + "epoch": 0.44384765625, + "grad_norm": 0.21029891073703766, + "learning_rate": 0.0003261065871628418, + "loss": 1.8036, + "step": 4545 + }, + { + "epoch": 0.4439453125, + "grad_norm": 0.22057075798511505, + "learning_rate": 0.0003260373308386604, + "loss": 1.8392, + "step": 4546 + }, + { + "epoch": 0.44404296875, + "grad_norm": 0.20081530511379242, + "learning_rate": 0.00032596806941630175, + "loss": 1.8567, + "step": 4547 + }, + { + "epoch": 0.444140625, + "grad_norm": 0.20286759734153748, + "learning_rate": 0.0003258988029026845, + "loss": 1.7656, + "step": 4548 + }, + { + "epoch": 0.44423828125, + "grad_norm": 0.22599783539772034, + "learning_rate": 0.0003258295313047279, + "loss": 1.8068, + "step": 4549 + }, + { + "epoch": 0.4443359375, + "grad_norm": 0.1798899620771408, + "learning_rate": 0.00032576025462935127, + "loss": 1.8433, + "step": 4550 + }, + { + "epoch": 0.44443359375, + "grad_norm": 0.2571936845779419, + "learning_rate": 0.00032569097288347496, + "loss": 1.8405, + "step": 4551 + }, + { + "epoch": 0.44453125, + "grad_norm": 0.194194957613945, + "learning_rate": 0.00032562168607401954, + "loss": 1.8379, + "step": 4552 + }, + { + "epoch": 0.44462890625, + "grad_norm": 0.21969552338123322, + "learning_rate": 0.0003255523942079062, + "loss": 1.863, + "step": 4553 + }, + { + "epoch": 0.4447265625, + "grad_norm": 0.2289535403251648, + "learning_rate": 0.0003254830972920564, + "loss": 1.8698, + "step": 4554 + }, + { + "epoch": 0.44482421875, + "grad_norm": 0.20062187314033508, + "learning_rate": 0.0003254137953333926, + "loss": 1.8851, + "step": 4555 + }, + { + "epoch": 0.444921875, + "grad_norm": 0.26987403631210327, + "learning_rate": 0.0003253444883388371, + "loss": 1.8601, + "step": 4556 + }, + { + "epoch": 0.44501953125, + "grad_norm": 0.20700135827064514, + "learning_rate": 0.0003252751763153133, + "loss": 1.8286, + "step": 4557 + }, + { + "epoch": 0.4451171875, + "grad_norm": 0.275177001953125, + "learning_rate": 0.0003252058592697448, + "loss": 1.8307, + "step": 4558 + }, + { + "epoch": 0.44521484375, + "grad_norm": 0.25656774640083313, + "learning_rate": 0.0003251365372090558, + "loss": 1.8535, + "step": 4559 + }, + { + "epoch": 0.4453125, + "grad_norm": 0.20680679380893707, + "learning_rate": 0.00032506721014017074, + "loss": 1.8538, + "step": 4560 + }, + { + "epoch": 0.44541015625, + "grad_norm": 0.21911300718784332, + "learning_rate": 0.0003249978780700151, + "loss": 1.8574, + "step": 4561 + }, + { + "epoch": 0.4455078125, + "grad_norm": 0.26287397742271423, + "learning_rate": 0.0003249285410055143, + "loss": 1.8358, + "step": 4562 + }, + { + "epoch": 0.44560546875, + "grad_norm": 0.2139604240655899, + "learning_rate": 0.0003248591989535946, + "loss": 1.8372, + "step": 4563 + }, + { + "epoch": 0.445703125, + "grad_norm": 0.22483497858047485, + "learning_rate": 0.0003247898519211826, + "loss": 1.8617, + "step": 4564 + }, + { + "epoch": 0.44580078125, + "grad_norm": 0.23549950122833252, + "learning_rate": 0.0003247204999152055, + "loss": 1.7976, + "step": 4565 + }, + { + "epoch": 0.4458984375, + "grad_norm": 0.2233511060476303, + "learning_rate": 0.000324651142942591, + "loss": 1.8178, + "step": 4566 + }, + { + "epoch": 0.44599609375, + "grad_norm": 0.23170800507068634, + "learning_rate": 0.00032458178101026715, + "loss": 1.826, + "step": 4567 + }, + { + "epoch": 0.44609375, + "grad_norm": 0.2332051545381546, + "learning_rate": 0.0003245124141251626, + "loss": 1.7838, + "step": 4568 + }, + { + "epoch": 0.44619140625, + "grad_norm": 0.2292395979166031, + "learning_rate": 0.0003244430422942066, + "loss": 1.8636, + "step": 4569 + }, + { + "epoch": 0.4462890625, + "grad_norm": 0.19580964744091034, + "learning_rate": 0.00032437366552432866, + "loss": 1.8412, + "step": 4570 + }, + { + "epoch": 0.44638671875, + "grad_norm": 0.22505177557468414, + "learning_rate": 0.0003243042838224589, + "loss": 1.8197, + "step": 4571 + }, + { + "epoch": 0.446484375, + "grad_norm": 0.18000158667564392, + "learning_rate": 0.000324234897195528, + "loss": 1.8399, + "step": 4572 + }, + { + "epoch": 0.44658203125, + "grad_norm": 0.2370178997516632, + "learning_rate": 0.00032416550565046715, + "loss": 1.842, + "step": 4573 + }, + { + "epoch": 0.4466796875, + "grad_norm": 0.387603223323822, + "learning_rate": 0.0003240961091942078, + "loss": 1.8703, + "step": 4574 + }, + { + "epoch": 0.44677734375, + "grad_norm": 0.2057318389415741, + "learning_rate": 0.000324026707833682, + "loss": 1.7813, + "step": 4575 + }, + { + "epoch": 0.446875, + "grad_norm": 0.26612409949302673, + "learning_rate": 0.0003239573015758225, + "loss": 1.8523, + "step": 4576 + }, + { + "epoch": 0.44697265625, + "grad_norm": 0.21448631584644318, + "learning_rate": 0.00032388789042756233, + "loss": 1.8147, + "step": 4577 + }, + { + "epoch": 0.4470703125, + "grad_norm": 0.22870376706123352, + "learning_rate": 0.00032381847439583503, + "loss": 1.8709, + "step": 4578 + }, + { + "epoch": 0.44716796875, + "grad_norm": 0.19087672233581543, + "learning_rate": 0.00032374905348757447, + "loss": 1.8498, + "step": 4579 + }, + { + "epoch": 0.447265625, + "grad_norm": 0.20512868463993073, + "learning_rate": 0.00032367962770971553, + "loss": 1.8191, + "step": 4580 + }, + { + "epoch": 0.44736328125, + "grad_norm": 0.22222061455249786, + "learning_rate": 0.000323610197069193, + "loss": 1.8537, + "step": 4581 + }, + { + "epoch": 0.4474609375, + "grad_norm": 0.249727264046669, + "learning_rate": 0.0003235407615729424, + "loss": 1.8624, + "step": 4582 + }, + { + "epoch": 0.44755859375, + "grad_norm": 0.19810056686401367, + "learning_rate": 0.00032347132122789973, + "loss": 1.8805, + "step": 4583 + }, + { + "epoch": 0.44765625, + "grad_norm": 0.28888627886772156, + "learning_rate": 0.0003234018760410015, + "loss": 1.8309, + "step": 4584 + }, + { + "epoch": 0.44775390625, + "grad_norm": 0.2219923883676529, + "learning_rate": 0.0003233324260191846, + "loss": 1.8829, + "step": 4585 + }, + { + "epoch": 0.4478515625, + "grad_norm": 0.2210731953382492, + "learning_rate": 0.00032326297116938666, + "loss": 1.85, + "step": 4586 + }, + { + "epoch": 0.44794921875, + "grad_norm": 0.2635856866836548, + "learning_rate": 0.00032319351149854544, + "loss": 1.8324, + "step": 4587 + }, + { + "epoch": 0.448046875, + "grad_norm": 0.19042371213436127, + "learning_rate": 0.0003231240470135993, + "loss": 1.8668, + "step": 4588 + }, + { + "epoch": 0.44814453125, + "grad_norm": 0.26712489128112793, + "learning_rate": 0.0003230545777214872, + "loss": 1.8891, + "step": 4589 + }, + { + "epoch": 0.4482421875, + "grad_norm": 0.20257070660591125, + "learning_rate": 0.0003229851036291486, + "loss": 1.822, + "step": 4590 + }, + { + "epoch": 0.44833984375, + "grad_norm": 0.24346715211868286, + "learning_rate": 0.00032291562474352314, + "loss": 1.845, + "step": 4591 + }, + { + "epoch": 0.4484375, + "grad_norm": 0.19787989556789398, + "learning_rate": 0.00032284614107155136, + "loss": 1.8411, + "step": 4592 + }, + { + "epoch": 0.44853515625, + "grad_norm": 0.20641331374645233, + "learning_rate": 0.000322776652620174, + "loss": 1.8279, + "step": 4593 + }, + { + "epoch": 0.4486328125, + "grad_norm": 0.1935081034898758, + "learning_rate": 0.0003227071593963322, + "loss": 1.8451, + "step": 4594 + }, + { + "epoch": 0.44873046875, + "grad_norm": 0.2476644665002823, + "learning_rate": 0.00032263766140696784, + "loss": 1.8205, + "step": 4595 + }, + { + "epoch": 0.448828125, + "grad_norm": 0.22297869622707367, + "learning_rate": 0.0003225681586590232, + "loss": 1.8599, + "step": 4596 + }, + { + "epoch": 0.44892578125, + "grad_norm": 0.2534295618534088, + "learning_rate": 0.0003224986511594408, + "loss": 1.8362, + "step": 4597 + }, + { + "epoch": 0.4490234375, + "grad_norm": 0.2747684121131897, + "learning_rate": 0.00032242913891516415, + "loss": 1.8534, + "step": 4598 + }, + { + "epoch": 0.44912109375, + "grad_norm": 0.2065267711877823, + "learning_rate": 0.00032235962193313654, + "loss": 1.8454, + "step": 4599 + }, + { + "epoch": 0.44921875, + "grad_norm": 0.23845352232456207, + "learning_rate": 0.0003222901002203023, + "loss": 1.8486, + "step": 4600 + }, + { + "epoch": 0.44931640625, + "grad_norm": 0.2285664677619934, + "learning_rate": 0.0003222205737836061, + "loss": 1.8862, + "step": 4601 + }, + { + "epoch": 0.4494140625, + "grad_norm": 0.199177548289299, + "learning_rate": 0.0003221510426299928, + "loss": 1.8045, + "step": 4602 + }, + { + "epoch": 0.44951171875, + "grad_norm": 0.21812167763710022, + "learning_rate": 0.00032208150676640813, + "loss": 1.8548, + "step": 4603 + }, + { + "epoch": 0.449609375, + "grad_norm": 0.19792795181274414, + "learning_rate": 0.000322011966199798, + "loss": 1.8192, + "step": 4604 + }, + { + "epoch": 0.44970703125, + "grad_norm": 0.2088000774383545, + "learning_rate": 0.000321942420937109, + "loss": 1.8137, + "step": 4605 + }, + { + "epoch": 0.4498046875, + "grad_norm": 0.1992752104997635, + "learning_rate": 0.00032187287098528786, + "loss": 1.8331, + "step": 4606 + }, + { + "epoch": 0.44990234375, + "grad_norm": 0.24328278005123138, + "learning_rate": 0.00032180331635128236, + "loss": 1.91, + "step": 4607 + }, + { + "epoch": 0.45, + "grad_norm": 0.21111084520816803, + "learning_rate": 0.0003217337570420401, + "loss": 1.8614, + "step": 4608 + }, + { + "epoch": 0.45009765625, + "grad_norm": 0.22330433130264282, + "learning_rate": 0.0003216641930645095, + "loss": 1.8478, + "step": 4609 + }, + { + "epoch": 0.4501953125, + "grad_norm": 0.22892868518829346, + "learning_rate": 0.0003215946244256395, + "loss": 1.856, + "step": 4610 + }, + { + "epoch": 0.45029296875, + "grad_norm": 0.18284457921981812, + "learning_rate": 0.0003215250511323792, + "loss": 1.8035, + "step": 4611 + }, + { + "epoch": 0.450390625, + "grad_norm": 0.21104735136032104, + "learning_rate": 0.0003214554731916786, + "loss": 1.814, + "step": 4612 + }, + { + "epoch": 0.45048828125, + "grad_norm": 0.1928839236497879, + "learning_rate": 0.00032138589061048774, + "loss": 1.8403, + "step": 4613 + }, + { + "epoch": 0.4505859375, + "grad_norm": 0.2700890004634857, + "learning_rate": 0.00032131630339575734, + "loss": 1.8487, + "step": 4614 + }, + { + "epoch": 0.45068359375, + "grad_norm": 0.25254520773887634, + "learning_rate": 0.0003212467115544386, + "loss": 1.8956, + "step": 4615 + }, + { + "epoch": 0.45078125, + "grad_norm": 0.19750279188156128, + "learning_rate": 0.00032117711509348294, + "loss": 1.8706, + "step": 4616 + }, + { + "epoch": 0.45087890625, + "grad_norm": 0.2669914662837982, + "learning_rate": 0.0003211075140198427, + "loss": 1.8755, + "step": 4617 + }, + { + "epoch": 0.4509765625, + "grad_norm": 0.20006148517131805, + "learning_rate": 0.0003210379083404703, + "loss": 1.8485, + "step": 4618 + }, + { + "epoch": 0.45107421875, + "grad_norm": 0.22197791934013367, + "learning_rate": 0.0003209682980623186, + "loss": 1.8702, + "step": 4619 + }, + { + "epoch": 0.451171875, + "grad_norm": 0.23509272933006287, + "learning_rate": 0.00032089868319234126, + "loss": 1.8208, + "step": 4620 + }, + { + "epoch": 0.45126953125, + "grad_norm": 0.20819897949695587, + "learning_rate": 0.0003208290637374921, + "loss": 1.8235, + "step": 4621 + }, + { + "epoch": 0.4513671875, + "grad_norm": 0.2182103544473648, + "learning_rate": 0.0003207594397047255, + "loss": 1.8122, + "step": 4622 + }, + { + "epoch": 0.45146484375, + "grad_norm": 0.2312661111354828, + "learning_rate": 0.00032068981110099613, + "loss": 1.8667, + "step": 4623 + }, + { + "epoch": 0.4515625, + "grad_norm": 0.22262756526470184, + "learning_rate": 0.0003206201779332596, + "loss": 1.8487, + "step": 4624 + }, + { + "epoch": 0.45166015625, + "grad_norm": 0.19101223349571228, + "learning_rate": 0.0003205505402084713, + "loss": 1.858, + "step": 4625 + }, + { + "epoch": 0.4517578125, + "grad_norm": 0.23975621163845062, + "learning_rate": 0.0003204808979335877, + "loss": 1.8766, + "step": 4626 + }, + { + "epoch": 0.45185546875, + "grad_norm": 0.1717529445886612, + "learning_rate": 0.0003204112511155652, + "loss": 1.8459, + "step": 4627 + }, + { + "epoch": 0.451953125, + "grad_norm": 0.2034163922071457, + "learning_rate": 0.00032034159976136105, + "loss": 1.8183, + "step": 4628 + }, + { + "epoch": 0.45205078125, + "grad_norm": 0.2004881650209427, + "learning_rate": 0.00032027194387793285, + "loss": 1.8368, + "step": 4629 + }, + { + "epoch": 0.4521484375, + "grad_norm": 0.2012924998998642, + "learning_rate": 0.0003202022834722384, + "loss": 1.8415, + "step": 4630 + }, + { + "epoch": 0.45224609375, + "grad_norm": 0.21484839916229248, + "learning_rate": 0.00032013261855123637, + "loss": 1.8123, + "step": 4631 + }, + { + "epoch": 0.45234375, + "grad_norm": 0.18154776096343994, + "learning_rate": 0.0003200629491218856, + "loss": 1.864, + "step": 4632 + }, + { + "epoch": 0.45244140625, + "grad_norm": 0.20559240877628326, + "learning_rate": 0.0003199932751911454, + "loss": 1.8732, + "step": 4633 + }, + { + "epoch": 0.4525390625, + "grad_norm": 0.22608701884746552, + "learning_rate": 0.0003199235967659756, + "loss": 1.8225, + "step": 4634 + }, + { + "epoch": 0.45263671875, + "grad_norm": 0.22271795570850372, + "learning_rate": 0.0003198539138533365, + "loss": 1.833, + "step": 4635 + }, + { + "epoch": 0.452734375, + "grad_norm": 0.21005083620548248, + "learning_rate": 0.00031978422646018886, + "loss": 1.8308, + "step": 4636 + }, + { + "epoch": 0.45283203125, + "grad_norm": 0.22713293135166168, + "learning_rate": 0.0003197145345934935, + "loss": 1.8179, + "step": 4637 + }, + { + "epoch": 0.4529296875, + "grad_norm": 0.2946507930755615, + "learning_rate": 0.0003196448382602125, + "loss": 1.8665, + "step": 4638 + }, + { + "epoch": 0.45302734375, + "grad_norm": 0.24628376960754395, + "learning_rate": 0.0003195751374673076, + "loss": 1.8204, + "step": 4639 + }, + { + "epoch": 0.453125, + "grad_norm": 0.2525516152381897, + "learning_rate": 0.0003195054322217414, + "loss": 1.8574, + "step": 4640 + }, + { + "epoch": 0.45322265625, + "grad_norm": 0.2724788188934326, + "learning_rate": 0.0003194357225304768, + "loss": 1.8386, + "step": 4641 + }, + { + "epoch": 0.4533203125, + "grad_norm": 0.24351954460144043, + "learning_rate": 0.00031936600840047706, + "loss": 1.8179, + "step": 4642 + }, + { + "epoch": 0.45341796875, + "grad_norm": 0.23132874071598053, + "learning_rate": 0.0003192962898387063, + "loss": 1.8607, + "step": 4643 + }, + { + "epoch": 0.453515625, + "grad_norm": 0.24049578607082367, + "learning_rate": 0.00031922656685212853, + "loss": 1.8435, + "step": 4644 + }, + { + "epoch": 0.45361328125, + "grad_norm": 0.19537390768527985, + "learning_rate": 0.0003191568394477086, + "loss": 1.8654, + "step": 4645 + }, + { + "epoch": 0.4537109375, + "grad_norm": 0.22461603581905365, + "learning_rate": 0.00031908710763241157, + "loss": 1.8072, + "step": 4646 + }, + { + "epoch": 0.45380859375, + "grad_norm": 0.1938980519771576, + "learning_rate": 0.00031901737141320303, + "loss": 1.8501, + "step": 4647 + }, + { + "epoch": 0.45390625, + "grad_norm": 0.1806454211473465, + "learning_rate": 0.000318947630797049, + "loss": 1.8417, + "step": 4648 + }, + { + "epoch": 0.45400390625, + "grad_norm": 0.18523812294006348, + "learning_rate": 0.0003188778857909161, + "loss": 1.8341, + "step": 4649 + }, + { + "epoch": 0.4541015625, + "grad_norm": 0.18443694710731506, + "learning_rate": 0.0003188081364017711, + "loss": 1.8724, + "step": 4650 + }, + { + "epoch": 0.45419921875, + "grad_norm": 0.201009139418602, + "learning_rate": 0.0003187383826365813, + "loss": 1.8353, + "step": 4651 + }, + { + "epoch": 0.454296875, + "grad_norm": 0.18479257822036743, + "learning_rate": 0.0003186686245023147, + "loss": 1.8154, + "step": 4652 + }, + { + "epoch": 0.45439453125, + "grad_norm": 0.22350969910621643, + "learning_rate": 0.00031859886200593935, + "loss": 1.8358, + "step": 4653 + }, + { + "epoch": 0.4544921875, + "grad_norm": 0.19744180142879486, + "learning_rate": 0.00031852909515442385, + "loss": 1.8177, + "step": 4654 + }, + { + "epoch": 0.45458984375, + "grad_norm": 0.22612422704696655, + "learning_rate": 0.0003184593239547375, + "loss": 1.8113, + "step": 4655 + }, + { + "epoch": 0.4546875, + "grad_norm": 0.21232947707176208, + "learning_rate": 0.0003183895484138496, + "loss": 1.8488, + "step": 4656 + }, + { + "epoch": 0.45478515625, + "grad_norm": 0.1647469848394394, + "learning_rate": 0.0003183197685387303, + "loss": 1.8376, + "step": 4657 + }, + { + "epoch": 0.4548828125, + "grad_norm": 0.20144745707511902, + "learning_rate": 0.00031824998433634973, + "loss": 1.824, + "step": 4658 + }, + { + "epoch": 0.45498046875, + "grad_norm": 0.1921469271183014, + "learning_rate": 0.00031818019581367907, + "loss": 1.8716, + "step": 4659 + }, + { + "epoch": 0.455078125, + "grad_norm": 0.21708911657333374, + "learning_rate": 0.0003181104029776893, + "loss": 1.8351, + "step": 4660 + }, + { + "epoch": 0.45517578125, + "grad_norm": 0.18313667178153992, + "learning_rate": 0.00031804060583535217, + "loss": 1.7719, + "step": 4661 + }, + { + "epoch": 0.4552734375, + "grad_norm": 0.21659673750400543, + "learning_rate": 0.00031797080439363996, + "loss": 1.8509, + "step": 4662 + }, + { + "epoch": 0.45537109375, + "grad_norm": 0.21164323389530182, + "learning_rate": 0.00031790099865952495, + "loss": 1.8348, + "step": 4663 + }, + { + "epoch": 0.45546875, + "grad_norm": 0.19318445026874542, + "learning_rate": 0.0003178311886399803, + "loss": 1.8157, + "step": 4664 + }, + { + "epoch": 0.45556640625, + "grad_norm": 0.2079407125711441, + "learning_rate": 0.00031776137434197924, + "loss": 1.8298, + "step": 4665 + }, + { + "epoch": 0.4556640625, + "grad_norm": 0.23482640087604523, + "learning_rate": 0.0003176915557724959, + "loss": 1.8816, + "step": 4666 + }, + { + "epoch": 0.45576171875, + "grad_norm": 0.2438378483057022, + "learning_rate": 0.00031762173293850424, + "loss": 1.8628, + "step": 4667 + }, + { + "epoch": 0.455859375, + "grad_norm": 0.2278955578804016, + "learning_rate": 0.0003175519058469791, + "loss": 1.8159, + "step": 4668 + }, + { + "epoch": 0.45595703125, + "grad_norm": 0.2014375478029251, + "learning_rate": 0.0003174820745048955, + "loss": 1.887, + "step": 4669 + }, + { + "epoch": 0.4560546875, + "grad_norm": 0.24122567474842072, + "learning_rate": 0.00031741223891922906, + "loss": 1.8403, + "step": 4670 + }, + { + "epoch": 0.45615234375, + "grad_norm": 0.1940990686416626, + "learning_rate": 0.00031734239909695565, + "loss": 1.8293, + "step": 4671 + }, + { + "epoch": 0.45625, + "grad_norm": 0.21825401484966278, + "learning_rate": 0.00031727255504505165, + "loss": 1.8123, + "step": 4672 + }, + { + "epoch": 0.45634765625, + "grad_norm": 0.22472091019153595, + "learning_rate": 0.00031720270677049395, + "loss": 1.8572, + "step": 4673 + }, + { + "epoch": 0.4564453125, + "grad_norm": 0.2011629343032837, + "learning_rate": 0.00031713285428025973, + "loss": 1.8251, + "step": 4674 + }, + { + "epoch": 0.45654296875, + "grad_norm": 0.19883796572685242, + "learning_rate": 0.00031706299758132663, + "loss": 1.8322, + "step": 4675 + }, + { + "epoch": 0.456640625, + "grad_norm": 0.20913474261760712, + "learning_rate": 0.00031699313668067275, + "loss": 1.8519, + "step": 4676 + }, + { + "epoch": 0.45673828125, + "grad_norm": 0.19562242925167084, + "learning_rate": 0.0003169232715852765, + "loss": 1.742, + "step": 4677 + }, + { + "epoch": 0.4568359375, + "grad_norm": 0.19825175404548645, + "learning_rate": 0.0003168534023021169, + "loss": 1.8201, + "step": 4678 + }, + { + "epoch": 0.45693359375, + "grad_norm": 0.19559837877750397, + "learning_rate": 0.000316783528838173, + "loss": 1.8539, + "step": 4679 + }, + { + "epoch": 0.45703125, + "grad_norm": 0.2345285564661026, + "learning_rate": 0.00031671365120042495, + "loss": 1.7824, + "step": 4680 + }, + { + "epoch": 0.45712890625, + "grad_norm": 0.26567429304122925, + "learning_rate": 0.00031664376939585255, + "loss": 1.8448, + "step": 4681 + }, + { + "epoch": 0.4572265625, + "grad_norm": 0.24556522071361542, + "learning_rate": 0.00031657388343143657, + "loss": 1.8422, + "step": 4682 + }, + { + "epoch": 0.45732421875, + "grad_norm": 0.19742777943611145, + "learning_rate": 0.000316503993314158, + "loss": 1.8493, + "step": 4683 + }, + { + "epoch": 0.457421875, + "grad_norm": 0.2256293147802353, + "learning_rate": 0.0003164340990509982, + "loss": 1.8233, + "step": 4684 + }, + { + "epoch": 0.45751953125, + "grad_norm": 0.2693006098270416, + "learning_rate": 0.0003163642006489389, + "loss": 1.9092, + "step": 4685 + }, + { + "epoch": 0.4576171875, + "grad_norm": 0.19180917739868164, + "learning_rate": 0.00031629429811496235, + "loss": 1.8163, + "step": 4686 + }, + { + "epoch": 0.45771484375, + "grad_norm": 0.23703311383724213, + "learning_rate": 0.0003162243914560514, + "loss": 1.8646, + "step": 4687 + }, + { + "epoch": 0.4578125, + "grad_norm": 0.22602078318595886, + "learning_rate": 0.0003161544806791889, + "loss": 1.8588, + "step": 4688 + }, + { + "epoch": 0.45791015625, + "grad_norm": 0.22127455472946167, + "learning_rate": 0.0003160845657913584, + "loss": 1.8436, + "step": 4689 + }, + { + "epoch": 0.4580078125, + "grad_norm": 0.25368431210517883, + "learning_rate": 0.0003160146467995437, + "loss": 1.803, + "step": 4690 + }, + { + "epoch": 0.45810546875, + "grad_norm": 0.21059094369411469, + "learning_rate": 0.00031594472371072917, + "loss": 1.8458, + "step": 4691 + }, + { + "epoch": 0.458203125, + "grad_norm": 0.19106703996658325, + "learning_rate": 0.0003158747965318995, + "loss": 1.8406, + "step": 4692 + }, + { + "epoch": 0.45830078125, + "grad_norm": 0.20046930015087128, + "learning_rate": 0.0003158048652700397, + "loss": 1.8376, + "step": 4693 + }, + { + "epoch": 0.4583984375, + "grad_norm": 0.17693589627742767, + "learning_rate": 0.0003157349299321355, + "loss": 1.8098, + "step": 4694 + }, + { + "epoch": 0.45849609375, + "grad_norm": 0.19985699653625488, + "learning_rate": 0.0003156649905251725, + "loss": 1.8566, + "step": 4695 + }, + { + "epoch": 0.45859375, + "grad_norm": 0.21398624777793884, + "learning_rate": 0.0003155950470561373, + "loss": 1.86, + "step": 4696 + }, + { + "epoch": 0.45869140625, + "grad_norm": 0.19084253907203674, + "learning_rate": 0.00031552509953201657, + "loss": 1.847, + "step": 4697 + }, + { + "epoch": 0.4587890625, + "grad_norm": 0.1904585063457489, + "learning_rate": 0.0003154551479597974, + "loss": 1.8069, + "step": 4698 + }, + { + "epoch": 0.45888671875, + "grad_norm": 0.20789475739002228, + "learning_rate": 0.0003153851923464673, + "loss": 1.8255, + "step": 4699 + }, + { + "epoch": 0.458984375, + "grad_norm": 0.19003227353096008, + "learning_rate": 0.0003153152326990143, + "loss": 1.8396, + "step": 4700 + }, + { + "epoch": 0.45908203125, + "grad_norm": 0.17602834105491638, + "learning_rate": 0.00031524526902442667, + "loss": 1.8203, + "step": 4701 + }, + { + "epoch": 0.4591796875, + "grad_norm": 0.2242649793624878, + "learning_rate": 0.00031517530132969324, + "loss": 1.8263, + "step": 4702 + }, + { + "epoch": 0.45927734375, + "grad_norm": 0.21256732940673828, + "learning_rate": 0.00031510532962180315, + "loss": 1.8385, + "step": 4703 + }, + { + "epoch": 0.459375, + "grad_norm": 0.23991239070892334, + "learning_rate": 0.0003150353539077459, + "loss": 1.7929, + "step": 4704 + }, + { + "epoch": 0.45947265625, + "grad_norm": 0.21919970214366913, + "learning_rate": 0.0003149653741945115, + "loss": 1.8649, + "step": 4705 + }, + { + "epoch": 0.4595703125, + "grad_norm": 0.20121543109416962, + "learning_rate": 0.0003148953904890903, + "loss": 1.8084, + "step": 4706 + }, + { + "epoch": 0.45966796875, + "grad_norm": 0.2023702710866928, + "learning_rate": 0.00031482540279847293, + "loss": 1.8375, + "step": 4707 + }, + { + "epoch": 0.459765625, + "grad_norm": 0.20311948657035828, + "learning_rate": 0.0003147554111296508, + "loss": 1.8536, + "step": 4708 + }, + { + "epoch": 0.45986328125, + "grad_norm": 0.19386813044548035, + "learning_rate": 0.00031468541548961514, + "loss": 1.8395, + "step": 4709 + }, + { + "epoch": 0.4599609375, + "grad_norm": 0.1801415979862213, + "learning_rate": 0.00031461541588535813, + "loss": 1.8171, + "step": 4710 + }, + { + "epoch": 0.46005859375, + "grad_norm": 0.22196736931800842, + "learning_rate": 0.000314545412323872, + "loss": 1.8151, + "step": 4711 + }, + { + "epoch": 0.46015625, + "grad_norm": 0.21495920419692993, + "learning_rate": 0.00031447540481214966, + "loss": 1.7585, + "step": 4712 + }, + { + "epoch": 0.46025390625, + "grad_norm": 0.21375064551830292, + "learning_rate": 0.0003144053933571839, + "loss": 1.8212, + "step": 4713 + }, + { + "epoch": 0.4603515625, + "grad_norm": 0.23266538977622986, + "learning_rate": 0.0003143353779659685, + "loss": 1.8824, + "step": 4714 + }, + { + "epoch": 0.46044921875, + "grad_norm": 0.20703117549419403, + "learning_rate": 0.00031426535864549746, + "loss": 1.8231, + "step": 4715 + }, + { + "epoch": 0.460546875, + "grad_norm": 0.19576159119606018, + "learning_rate": 0.0003141953354027648, + "loss": 1.7841, + "step": 4716 + }, + { + "epoch": 0.46064453125, + "grad_norm": 0.23798930644989014, + "learning_rate": 0.00031412530824476544, + "loss": 1.8671, + "step": 4717 + }, + { + "epoch": 0.4607421875, + "grad_norm": 0.17755867540836334, + "learning_rate": 0.0003140552771784943, + "loss": 1.7874, + "step": 4718 + }, + { + "epoch": 0.46083984375, + "grad_norm": 0.22959421575069427, + "learning_rate": 0.0003139852422109471, + "loss": 1.7994, + "step": 4719 + }, + { + "epoch": 0.4609375, + "grad_norm": 0.1911461055278778, + "learning_rate": 0.00031391520334911956, + "loss": 1.8492, + "step": 4720 + }, + { + "epoch": 0.46103515625, + "grad_norm": 0.20863354206085205, + "learning_rate": 0.00031384516060000794, + "loss": 1.8397, + "step": 4721 + }, + { + "epoch": 0.4611328125, + "grad_norm": 0.21101114153862, + "learning_rate": 0.0003137751139706089, + "loss": 1.8613, + "step": 4722 + }, + { + "epoch": 0.46123046875, + "grad_norm": 0.17822936177253723, + "learning_rate": 0.0003137050634679195, + "loss": 1.8547, + "step": 4723 + }, + { + "epoch": 0.461328125, + "grad_norm": 0.18411372601985931, + "learning_rate": 0.00031363500909893714, + "loss": 1.7851, + "step": 4724 + }, + { + "epoch": 0.46142578125, + "grad_norm": 0.2380722016096115, + "learning_rate": 0.00031356495087065967, + "loss": 1.8754, + "step": 4725 + }, + { + "epoch": 0.4615234375, + "grad_norm": 0.1815214604139328, + "learning_rate": 0.00031349488879008525, + "loss": 1.8613, + "step": 4726 + }, + { + "epoch": 0.46162109375, + "grad_norm": 0.2026418000459671, + "learning_rate": 0.00031342482286421245, + "loss": 1.8119, + "step": 4727 + }, + { + "epoch": 0.46171875, + "grad_norm": 0.201633021235466, + "learning_rate": 0.0003133547531000403, + "loss": 1.8222, + "step": 4728 + }, + { + "epoch": 0.46181640625, + "grad_norm": 0.20121082663536072, + "learning_rate": 0.0003132846795045681, + "loss": 1.8296, + "step": 4729 + }, + { + "epoch": 0.4619140625, + "grad_norm": 0.20398394763469696, + "learning_rate": 0.0003132146020847956, + "loss": 1.8039, + "step": 4730 + }, + { + "epoch": 0.46201171875, + "grad_norm": 0.23014071583747864, + "learning_rate": 0.00031314452084772287, + "loss": 1.8408, + "step": 4731 + }, + { + "epoch": 0.462109375, + "grad_norm": 0.18318094313144684, + "learning_rate": 0.00031307443580035044, + "loss": 1.847, + "step": 4732 + }, + { + "epoch": 0.46220703125, + "grad_norm": 0.24769657850265503, + "learning_rate": 0.0003130043469496793, + "loss": 1.7926, + "step": 4733 + }, + { + "epoch": 0.4623046875, + "grad_norm": 0.23016546666622162, + "learning_rate": 0.0003129342543027104, + "loss": 1.829, + "step": 4734 + }, + { + "epoch": 0.46240234375, + "grad_norm": 0.20693141222000122, + "learning_rate": 0.00031286415786644566, + "loss": 1.8134, + "step": 4735 + }, + { + "epoch": 0.4625, + "grad_norm": 0.2622755467891693, + "learning_rate": 0.000312794057647887, + "loss": 1.7994, + "step": 4736 + }, + { + "epoch": 0.46259765625, + "grad_norm": 0.2551204562187195, + "learning_rate": 0.00031272395365403676, + "loss": 1.8068, + "step": 4737 + }, + { + "epoch": 0.4626953125, + "grad_norm": 0.2194601446390152, + "learning_rate": 0.00031265384589189784, + "loss": 1.8373, + "step": 4738 + }, + { + "epoch": 0.46279296875, + "grad_norm": 0.2409636527299881, + "learning_rate": 0.0003125837343684732, + "loss": 1.8253, + "step": 4739 + }, + { + "epoch": 0.462890625, + "grad_norm": 0.18212181329727173, + "learning_rate": 0.00031251361909076655, + "loss": 1.8611, + "step": 4740 + }, + { + "epoch": 0.46298828125, + "grad_norm": 0.2114643007516861, + "learning_rate": 0.0003124435000657817, + "loss": 1.8038, + "step": 4741 + }, + { + "epoch": 0.4630859375, + "grad_norm": 0.18933500349521637, + "learning_rate": 0.0003123733773005228, + "loss": 1.8489, + "step": 4742 + }, + { + "epoch": 0.46318359375, + "grad_norm": 0.26138171553611755, + "learning_rate": 0.00031230325080199467, + "loss": 1.8637, + "step": 4743 + }, + { + "epoch": 0.46328125, + "grad_norm": 0.19251863658428192, + "learning_rate": 0.0003122331205772023, + "loss": 1.8781, + "step": 4744 + }, + { + "epoch": 0.46337890625, + "grad_norm": 0.22503794729709625, + "learning_rate": 0.0003121629866331509, + "loss": 1.8259, + "step": 4745 + }, + { + "epoch": 0.4634765625, + "grad_norm": 0.18829478323459625, + "learning_rate": 0.0003120928489768465, + "loss": 1.8363, + "step": 4746 + }, + { + "epoch": 0.46357421875, + "grad_norm": 0.23937945067882538, + "learning_rate": 0.00031202270761529504, + "loss": 1.8785, + "step": 4747 + }, + { + "epoch": 0.463671875, + "grad_norm": 0.19133053719997406, + "learning_rate": 0.0003119525625555031, + "loss": 1.7037, + "step": 4748 + }, + { + "epoch": 0.46376953125, + "grad_norm": 0.238962322473526, + "learning_rate": 0.00031188241380447753, + "loss": 1.8661, + "step": 4749 + }, + { + "epoch": 0.4638671875, + "grad_norm": 0.19993072748184204, + "learning_rate": 0.00031181226136922565, + "loss": 1.8326, + "step": 4750 + }, + { + "epoch": 0.46396484375, + "grad_norm": 0.22920852899551392, + "learning_rate": 0.00031174210525675486, + "loss": 1.8469, + "step": 4751 + }, + { + "epoch": 0.4640625, + "grad_norm": 0.1875825673341751, + "learning_rate": 0.0003116719454740733, + "loss": 1.8428, + "step": 4752 + }, + { + "epoch": 0.46416015625, + "grad_norm": 0.17680490016937256, + "learning_rate": 0.0003116017820281893, + "loss": 1.857, + "step": 4753 + }, + { + "epoch": 0.4642578125, + "grad_norm": 0.19791361689567566, + "learning_rate": 0.00031153161492611153, + "loss": 1.8623, + "step": 4754 + }, + { + "epoch": 0.46435546875, + "grad_norm": 0.19239719212055206, + "learning_rate": 0.00031146144417484905, + "loss": 1.8489, + "step": 4755 + }, + { + "epoch": 0.464453125, + "grad_norm": 0.25808653235435486, + "learning_rate": 0.0003113912697814113, + "loss": 1.8682, + "step": 4756 + }, + { + "epoch": 0.46455078125, + "grad_norm": 0.24941441416740417, + "learning_rate": 0.0003113210917528082, + "loss": 1.8667, + "step": 4757 + }, + { + "epoch": 0.4646484375, + "grad_norm": 0.22093206644058228, + "learning_rate": 0.00031125091009604974, + "loss": 1.8292, + "step": 4758 + }, + { + "epoch": 0.46474609375, + "grad_norm": 0.2532329559326172, + "learning_rate": 0.00031118072481814646, + "loss": 1.8661, + "step": 4759 + }, + { + "epoch": 0.46484375, + "grad_norm": 0.23447300493717194, + "learning_rate": 0.00031111053592610934, + "loss": 1.856, + "step": 4760 + }, + { + "epoch": 0.46494140625, + "grad_norm": 0.2655586898326874, + "learning_rate": 0.00031104034342694964, + "loss": 1.8637, + "step": 4761 + }, + { + "epoch": 0.4650390625, + "grad_norm": 0.24939312040805817, + "learning_rate": 0.00031097014732767885, + "loss": 1.856, + "step": 4762 + }, + { + "epoch": 0.46513671875, + "grad_norm": 0.22842450439929962, + "learning_rate": 0.00031089994763530904, + "loss": 1.8004, + "step": 4763 + }, + { + "epoch": 0.465234375, + "grad_norm": 0.23652945458889008, + "learning_rate": 0.00031082974435685257, + "loss": 1.8637, + "step": 4764 + }, + { + "epoch": 0.46533203125, + "grad_norm": 0.2600919306278229, + "learning_rate": 0.00031075953749932193, + "loss": 1.865, + "step": 4765 + }, + { + "epoch": 0.4654296875, + "grad_norm": 0.23312941193580627, + "learning_rate": 0.0003106893270697304, + "loss": 1.8382, + "step": 4766 + }, + { + "epoch": 0.46552734375, + "grad_norm": 0.22886423766613007, + "learning_rate": 0.0003106191130750912, + "loss": 1.8445, + "step": 4767 + }, + { + "epoch": 0.465625, + "grad_norm": 0.21378353238105774, + "learning_rate": 0.0003105488955224182, + "loss": 1.8366, + "step": 4768 + }, + { + "epoch": 0.46572265625, + "grad_norm": 0.20071300864219666, + "learning_rate": 0.0003104786744187255, + "loss": 1.8199, + "step": 4769 + }, + { + "epoch": 0.4658203125, + "grad_norm": 0.19779184460639954, + "learning_rate": 0.0003104084497710275, + "loss": 1.8652, + "step": 4770 + }, + { + "epoch": 0.46591796875, + "grad_norm": 0.22096706926822662, + "learning_rate": 0.00031033822158633913, + "loss": 1.8289, + "step": 4771 + }, + { + "epoch": 0.466015625, + "grad_norm": 0.1866249144077301, + "learning_rate": 0.0003102679898716754, + "loss": 1.8178, + "step": 4772 + }, + { + "epoch": 0.46611328125, + "grad_norm": 0.21741607785224915, + "learning_rate": 0.000310197754634052, + "loss": 1.8224, + "step": 4773 + }, + { + "epoch": 0.4662109375, + "grad_norm": 0.18589475750923157, + "learning_rate": 0.00031012751588048465, + "loss": 1.8504, + "step": 4774 + }, + { + "epoch": 0.46630859375, + "grad_norm": 0.18547822535037994, + "learning_rate": 0.00031005727361798973, + "loss": 1.8022, + "step": 4775 + }, + { + "epoch": 0.46640625, + "grad_norm": 0.22344110906124115, + "learning_rate": 0.0003099870278535837, + "loss": 1.8399, + "step": 4776 + }, + { + "epoch": 0.46650390625, + "grad_norm": 0.1753036230802536, + "learning_rate": 0.00030991677859428364, + "loss": 1.8744, + "step": 4777 + }, + { + "epoch": 0.4666015625, + "grad_norm": 0.20883294939994812, + "learning_rate": 0.00030984652584710665, + "loss": 1.8307, + "step": 4778 + }, + { + "epoch": 0.46669921875, + "grad_norm": 0.2109844982624054, + "learning_rate": 0.0003097762696190704, + "loss": 1.8164, + "step": 4779 + }, + { + "epoch": 0.466796875, + "grad_norm": 0.18941885232925415, + "learning_rate": 0.000309706009917193, + "loss": 1.8122, + "step": 4780 + }, + { + "epoch": 0.46689453125, + "grad_norm": 0.21899116039276123, + "learning_rate": 0.00030963574674849263, + "loss": 1.7913, + "step": 4781 + }, + { + "epoch": 0.4669921875, + "grad_norm": 0.2173784077167511, + "learning_rate": 0.0003095654801199881, + "loss": 1.8328, + "step": 4782 + }, + { + "epoch": 0.46708984375, + "grad_norm": 0.1898737996816635, + "learning_rate": 0.00030949521003869826, + "loss": 1.8587, + "step": 4783 + }, + { + "epoch": 0.4671875, + "grad_norm": 0.22946122288703918, + "learning_rate": 0.00030942493651164247, + "loss": 1.869, + "step": 4784 + }, + { + "epoch": 0.46728515625, + "grad_norm": 0.22063229978084564, + "learning_rate": 0.00030935465954584065, + "loss": 1.8588, + "step": 4785 + }, + { + "epoch": 0.4673828125, + "grad_norm": 0.24954596161842346, + "learning_rate": 0.0003092843791483125, + "loss": 1.8617, + "step": 4786 + }, + { + "epoch": 0.46748046875, + "grad_norm": 0.19222931563854218, + "learning_rate": 0.0003092140953260788, + "loss": 1.8434, + "step": 4787 + }, + { + "epoch": 0.467578125, + "grad_norm": 0.26410526037216187, + "learning_rate": 0.00030914380808615995, + "loss": 1.8371, + "step": 4788 + }, + { + "epoch": 0.46767578125, + "grad_norm": 0.208063542842865, + "learning_rate": 0.00030907351743557723, + "loss": 1.8176, + "step": 4789 + }, + { + "epoch": 0.4677734375, + "grad_norm": 0.21440909802913666, + "learning_rate": 0.00030900322338135196, + "loss": 1.8456, + "step": 4790 + }, + { + "epoch": 0.46787109375, + "grad_norm": 0.21355842053890228, + "learning_rate": 0.00030893292593050587, + "loss": 1.8031, + "step": 4791 + }, + { + "epoch": 0.46796875, + "grad_norm": 0.17706333100795746, + "learning_rate": 0.00030886262509006117, + "loss": 1.8279, + "step": 4792 + }, + { + "epoch": 0.46806640625, + "grad_norm": 0.18919691443443298, + "learning_rate": 0.00030879232086704014, + "loss": 1.8368, + "step": 4793 + }, + { + "epoch": 0.4681640625, + "grad_norm": 0.18476110696792603, + "learning_rate": 0.00030872201326846574, + "loss": 1.8686, + "step": 4794 + }, + { + "epoch": 0.46826171875, + "grad_norm": 0.20892632007598877, + "learning_rate": 0.00030865170230136085, + "loss": 1.8322, + "step": 4795 + }, + { + "epoch": 0.468359375, + "grad_norm": 0.23286119103431702, + "learning_rate": 0.00030858138797274913, + "loss": 1.8497, + "step": 4796 + }, + { + "epoch": 0.46845703125, + "grad_norm": 0.18783611059188843, + "learning_rate": 0.00030851107028965417, + "loss": 1.8145, + "step": 4797 + }, + { + "epoch": 0.4685546875, + "grad_norm": 0.2191592901945114, + "learning_rate": 0.00030844074925910023, + "loss": 1.8404, + "step": 4798 + }, + { + "epoch": 0.46865234375, + "grad_norm": 0.2329634130001068, + "learning_rate": 0.0003083704248881117, + "loss": 1.8082, + "step": 4799 + }, + { + "epoch": 0.46875, + "grad_norm": 0.24307815730571747, + "learning_rate": 0.0003083000971837133, + "loss": 1.8451, + "step": 4800 + }, + { + "epoch": 0.46884765625, + "grad_norm": 0.18854624032974243, + "learning_rate": 0.00030822976615293025, + "loss": 1.8779, + "step": 4801 + }, + { + "epoch": 0.4689453125, + "grad_norm": 0.20139099657535553, + "learning_rate": 0.00030815943180278793, + "loss": 1.7947, + "step": 4802 + }, + { + "epoch": 0.46904296875, + "grad_norm": 0.20154617726802826, + "learning_rate": 0.00030808909414031224, + "loss": 1.8244, + "step": 4803 + }, + { + "epoch": 0.469140625, + "grad_norm": 0.2227020114660263, + "learning_rate": 0.00030801875317252903, + "loss": 1.8179, + "step": 4804 + }, + { + "epoch": 0.46923828125, + "grad_norm": 0.19972631335258484, + "learning_rate": 0.0003079484089064649, + "loss": 1.8663, + "step": 4805 + }, + { + "epoch": 0.4693359375, + "grad_norm": 0.2464173436164856, + "learning_rate": 0.0003078780613491468, + "loss": 1.8387, + "step": 4806 + }, + { + "epoch": 0.46943359375, + "grad_norm": 0.16700857877731323, + "learning_rate": 0.00030780771050760143, + "loss": 1.8156, + "step": 4807 + }, + { + "epoch": 0.46953125, + "grad_norm": 0.27698126435279846, + "learning_rate": 0.0003077373563888565, + "loss": 1.8527, + "step": 4808 + }, + { + "epoch": 0.46962890625, + "grad_norm": 0.24625715613365173, + "learning_rate": 0.0003076669989999398, + "loss": 1.8459, + "step": 4809 + }, + { + "epoch": 0.4697265625, + "grad_norm": 0.20767800509929657, + "learning_rate": 0.00030759663834787915, + "loss": 1.8617, + "step": 4810 + }, + { + "epoch": 0.46982421875, + "grad_norm": 0.2153705656528473, + "learning_rate": 0.0003075262744397032, + "loss": 1.8623, + "step": 4811 + }, + { + "epoch": 0.469921875, + "grad_norm": 0.26322141289711, + "learning_rate": 0.00030745590728244057, + "loss": 1.8434, + "step": 4812 + }, + { + "epoch": 0.47001953125, + "grad_norm": 0.22390654683113098, + "learning_rate": 0.0003073855368831204, + "loss": 1.8188, + "step": 4813 + }, + { + "epoch": 0.4701171875, + "grad_norm": 0.22162260115146637, + "learning_rate": 0.0003073151632487719, + "loss": 1.8019, + "step": 4814 + }, + { + "epoch": 0.47021484375, + "grad_norm": 0.29895639419555664, + "learning_rate": 0.00030724478638642486, + "loss": 1.8382, + "step": 4815 + }, + { + "epoch": 0.4703125, + "grad_norm": 0.22144287824630737, + "learning_rate": 0.0003071744063031094, + "loss": 1.8087, + "step": 4816 + }, + { + "epoch": 0.47041015625, + "grad_norm": 0.21702681481838226, + "learning_rate": 0.00030710402300585584, + "loss": 1.8183, + "step": 4817 + }, + { + "epoch": 0.4705078125, + "grad_norm": 0.19565224647521973, + "learning_rate": 0.0003070336365016947, + "loss": 1.8004, + "step": 4818 + }, + { + "epoch": 0.47060546875, + "grad_norm": 0.21208736300468445, + "learning_rate": 0.00030696324679765707, + "loss": 1.8431, + "step": 4819 + }, + { + "epoch": 0.470703125, + "grad_norm": 0.17192751169204712, + "learning_rate": 0.0003068928539007743, + "loss": 1.8063, + "step": 4820 + }, + { + "epoch": 0.47080078125, + "grad_norm": 0.20137910544872284, + "learning_rate": 0.00030682245781807793, + "loss": 1.8505, + "step": 4821 + }, + { + "epoch": 0.4708984375, + "grad_norm": 0.20366063714027405, + "learning_rate": 0.0003067520585566, + "loss": 1.8078, + "step": 4822 + }, + { + "epoch": 0.47099609375, + "grad_norm": 0.20955240726470947, + "learning_rate": 0.0003066816561233727, + "loss": 1.8682, + "step": 4823 + }, + { + "epoch": 0.47109375, + "grad_norm": 0.21498572826385498, + "learning_rate": 0.0003066112505254286, + "loss": 1.8392, + "step": 4824 + }, + { + "epoch": 0.47119140625, + "grad_norm": 0.18648019433021545, + "learning_rate": 0.0003065408417698006, + "loss": 1.8007, + "step": 4825 + }, + { + "epoch": 0.4712890625, + "grad_norm": 0.20473703742027283, + "learning_rate": 0.00030647042986352204, + "loss": 1.7969, + "step": 4826 + }, + { + "epoch": 0.47138671875, + "grad_norm": 0.1886153668165207, + "learning_rate": 0.0003064000148136262, + "loss": 1.8533, + "step": 4827 + }, + { + "epoch": 0.471484375, + "grad_norm": 0.23058368265628815, + "learning_rate": 0.0003063295966271471, + "loss": 1.8437, + "step": 4828 + }, + { + "epoch": 0.47158203125, + "grad_norm": 0.20837700366973877, + "learning_rate": 0.0003062591753111189, + "loss": 1.8072, + "step": 4829 + }, + { + "epoch": 0.4716796875, + "grad_norm": 0.1911066621541977, + "learning_rate": 0.000306188750872576, + "loss": 1.8039, + "step": 4830 + }, + { + "epoch": 0.47177734375, + "grad_norm": 0.24263250827789307, + "learning_rate": 0.0003061183233185532, + "loss": 1.8207, + "step": 4831 + }, + { + "epoch": 0.471875, + "grad_norm": 0.23435233533382416, + "learning_rate": 0.00030604789265608554, + "loss": 1.8643, + "step": 4832 + }, + { + "epoch": 0.47197265625, + "grad_norm": 0.2540555000305176, + "learning_rate": 0.0003059774588922085, + "loss": 1.7786, + "step": 4833 + }, + { + "epoch": 0.4720703125, + "grad_norm": 0.20325538516044617, + "learning_rate": 0.00030590702203395784, + "loss": 1.8221, + "step": 4834 + }, + { + "epoch": 0.47216796875, + "grad_norm": 0.3231290876865387, + "learning_rate": 0.00030583658208836934, + "loss": 1.8664, + "step": 4835 + }, + { + "epoch": 0.472265625, + "grad_norm": 0.23043441772460938, + "learning_rate": 0.0003057661390624796, + "loss": 1.8649, + "step": 4836 + }, + { + "epoch": 0.47236328125, + "grad_norm": 0.2705766558647156, + "learning_rate": 0.000305695692963325, + "loss": 1.8615, + "step": 4837 + }, + { + "epoch": 0.4724609375, + "grad_norm": 0.2516499161720276, + "learning_rate": 0.00030562524379794273, + "loss": 1.8781, + "step": 4838 + }, + { + "epoch": 0.47255859375, + "grad_norm": 0.23899440467357635, + "learning_rate": 0.0003055547915733699, + "loss": 1.8527, + "step": 4839 + }, + { + "epoch": 0.47265625, + "grad_norm": 0.23780567944049835, + "learning_rate": 0.0003054843362966441, + "loss": 1.8686, + "step": 4840 + }, + { + "epoch": 0.47275390625, + "grad_norm": 0.2458800971508026, + "learning_rate": 0.0003054138779748031, + "loss": 1.8402, + "step": 4841 + }, + { + "epoch": 0.4728515625, + "grad_norm": 0.21160176396369934, + "learning_rate": 0.0003053434166148851, + "loss": 1.84, + "step": 4842 + }, + { + "epoch": 0.47294921875, + "grad_norm": 0.19955572485923767, + "learning_rate": 0.0003052729522239287, + "loss": 1.8212, + "step": 4843 + }, + { + "epoch": 0.473046875, + "grad_norm": 0.19997259974479675, + "learning_rate": 0.0003052024848089725, + "loss": 1.8435, + "step": 4844 + }, + { + "epoch": 0.47314453125, + "grad_norm": 0.21002988517284393, + "learning_rate": 0.0003051320143770557, + "loss": 1.8503, + "step": 4845 + }, + { + "epoch": 0.4732421875, + "grad_norm": 0.20021061599254608, + "learning_rate": 0.00030506154093521754, + "loss": 1.8227, + "step": 4846 + }, + { + "epoch": 0.47333984375, + "grad_norm": 0.20722252130508423, + "learning_rate": 0.0003049910644904978, + "loss": 1.8288, + "step": 4847 + }, + { + "epoch": 0.4734375, + "grad_norm": 0.23750610649585724, + "learning_rate": 0.00030492058504993637, + "loss": 1.8581, + "step": 4848 + }, + { + "epoch": 0.47353515625, + "grad_norm": 0.19082392752170563, + "learning_rate": 0.0003048501026205736, + "loss": 1.8576, + "step": 4849 + }, + { + "epoch": 0.4736328125, + "grad_norm": 0.23766285181045532, + "learning_rate": 0.0003047796172094501, + "loss": 1.8303, + "step": 4850 + }, + { + "epoch": 0.47373046875, + "grad_norm": 0.21583178639411926, + "learning_rate": 0.0003047091288236066, + "loss": 1.789, + "step": 4851 + }, + { + "epoch": 0.473828125, + "grad_norm": 0.21662573516368866, + "learning_rate": 0.0003046386374700843, + "loss": 1.8523, + "step": 4852 + }, + { + "epoch": 0.47392578125, + "grad_norm": 0.22070710361003876, + "learning_rate": 0.0003045681431559247, + "loss": 1.8599, + "step": 4853 + }, + { + "epoch": 0.4740234375, + "grad_norm": 0.24383734166622162, + "learning_rate": 0.0003044976458881696, + "loss": 1.8268, + "step": 4854 + }, + { + "epoch": 0.47412109375, + "grad_norm": 0.2057112157344818, + "learning_rate": 0.0003044271456738609, + "loss": 1.8288, + "step": 4855 + }, + { + "epoch": 0.47421875, + "grad_norm": 0.1968592405319214, + "learning_rate": 0.0003043566425200411, + "loss": 1.8348, + "step": 4856 + }, + { + "epoch": 0.47431640625, + "grad_norm": 0.2273697406053543, + "learning_rate": 0.0003042861364337529, + "loss": 1.8611, + "step": 4857 + }, + { + "epoch": 0.4744140625, + "grad_norm": 0.18474043905735016, + "learning_rate": 0.00030421562742203895, + "loss": 1.8141, + "step": 4858 + }, + { + "epoch": 0.47451171875, + "grad_norm": 0.23579570651054382, + "learning_rate": 0.00030414511549194275, + "loss": 1.8601, + "step": 4859 + }, + { + "epoch": 0.474609375, + "grad_norm": 0.21270345151424408, + "learning_rate": 0.0003040746006505077, + "loss": 1.8268, + "step": 4860 + }, + { + "epoch": 0.47470703125, + "grad_norm": 0.2242332249879837, + "learning_rate": 0.0003040040829047776, + "loss": 1.8241, + "step": 4861 + }, + { + "epoch": 0.4748046875, + "grad_norm": 0.23544490337371826, + "learning_rate": 0.0003039335622617966, + "loss": 1.8104, + "step": 4862 + }, + { + "epoch": 0.47490234375, + "grad_norm": 0.2175990343093872, + "learning_rate": 0.000303863038728609, + "loss": 1.8609, + "step": 4863 + }, + { + "epoch": 0.475, + "grad_norm": 0.21629612147808075, + "learning_rate": 0.00030379251231225956, + "loss": 1.8484, + "step": 4864 + }, + { + "epoch": 0.47509765625, + "grad_norm": 0.24267800152301788, + "learning_rate": 0.00030372198301979324, + "loss": 1.809, + "step": 4865 + }, + { + "epoch": 0.4751953125, + "grad_norm": 0.1799250841140747, + "learning_rate": 0.0003036514508582553, + "loss": 1.8257, + "step": 4866 + }, + { + "epoch": 0.47529296875, + "grad_norm": 0.25859299302101135, + "learning_rate": 0.00030358091583469114, + "loss": 1.8236, + "step": 4867 + }, + { + "epoch": 0.475390625, + "grad_norm": 0.18856181204319, + "learning_rate": 0.00030351037795614685, + "loss": 1.809, + "step": 4868 + }, + { + "epoch": 0.47548828125, + "grad_norm": 0.24719011783599854, + "learning_rate": 0.00030343983722966826, + "loss": 1.8375, + "step": 4869 + }, + { + "epoch": 0.4755859375, + "grad_norm": 0.23080545663833618, + "learning_rate": 0.00030336929366230194, + "loss": 1.9084, + "step": 4870 + }, + { + "epoch": 0.47568359375, + "grad_norm": 0.2338431179523468, + "learning_rate": 0.00030329874726109444, + "loss": 1.8331, + "step": 4871 + }, + { + "epoch": 0.47578125, + "grad_norm": 0.20135995745658875, + "learning_rate": 0.0003032281980330929, + "loss": 1.8057, + "step": 4872 + }, + { + "epoch": 0.47587890625, + "grad_norm": 0.21494583785533905, + "learning_rate": 0.0003031576459853444, + "loss": 1.8325, + "step": 4873 + }, + { + "epoch": 0.4759765625, + "grad_norm": 0.2113523930311203, + "learning_rate": 0.0003030870911248965, + "loss": 1.8472, + "step": 4874 + }, + { + "epoch": 0.47607421875, + "grad_norm": 0.18979957699775696, + "learning_rate": 0.0003030165334587971, + "loss": 1.8702, + "step": 4875 + }, + { + "epoch": 0.476171875, + "grad_norm": 0.22056496143341064, + "learning_rate": 0.0003029459729940941, + "loss": 1.8397, + "step": 4876 + }, + { + "epoch": 0.47626953125, + "grad_norm": 0.23806160688400269, + "learning_rate": 0.00030287540973783603, + "loss": 1.8916, + "step": 4877 + }, + { + "epoch": 0.4763671875, + "grad_norm": 0.21538598835468292, + "learning_rate": 0.00030280484369707157, + "loss": 1.845, + "step": 4878 + }, + { + "epoch": 0.47646484375, + "grad_norm": 0.21205362677574158, + "learning_rate": 0.0003027342748788494, + "loss": 1.8413, + "step": 4879 + }, + { + "epoch": 0.4765625, + "grad_norm": 0.23677507042884827, + "learning_rate": 0.0003026637032902189, + "loss": 1.8386, + "step": 4880 + }, + { + "epoch": 0.47666015625, + "grad_norm": 0.18932418525218964, + "learning_rate": 0.0003025931289382296, + "loss": 1.8216, + "step": 4881 + }, + { + "epoch": 0.4767578125, + "grad_norm": 0.21288220584392548, + "learning_rate": 0.0003025225518299312, + "loss": 1.8326, + "step": 4882 + }, + { + "epoch": 0.47685546875, + "grad_norm": 0.19770745933055878, + "learning_rate": 0.0003024519719723736, + "loss": 1.8327, + "step": 4883 + }, + { + "epoch": 0.476953125, + "grad_norm": 0.19045844674110413, + "learning_rate": 0.00030238138937260724, + "loss": 1.8104, + "step": 4884 + }, + { + "epoch": 0.47705078125, + "grad_norm": 0.18676099181175232, + "learning_rate": 0.0003023108040376827, + "loss": 1.8833, + "step": 4885 + }, + { + "epoch": 0.4771484375, + "grad_norm": 0.17938444018363953, + "learning_rate": 0.0003022402159746507, + "loss": 1.7904, + "step": 4886 + }, + { + "epoch": 0.47724609375, + "grad_norm": 0.2277635782957077, + "learning_rate": 0.0003021696251905625, + "loss": 1.8592, + "step": 4887 + }, + { + "epoch": 0.47734375, + "grad_norm": 0.22101110219955444, + "learning_rate": 0.00030209903169246946, + "loss": 1.8322, + "step": 4888 + }, + { + "epoch": 0.47744140625, + "grad_norm": 0.20787768065929413, + "learning_rate": 0.0003020284354874233, + "loss": 1.8174, + "step": 4889 + }, + { + "epoch": 0.4775390625, + "grad_norm": 0.21741506457328796, + "learning_rate": 0.00030195783658247565, + "loss": 1.858, + "step": 4890 + }, + { + "epoch": 0.47763671875, + "grad_norm": 0.20655880868434906, + "learning_rate": 0.0003018872349846791, + "loss": 1.8527, + "step": 4891 + }, + { + "epoch": 0.477734375, + "grad_norm": 0.25105416774749756, + "learning_rate": 0.00030181663070108603, + "loss": 1.8398, + "step": 4892 + }, + { + "epoch": 0.47783203125, + "grad_norm": 0.23866187036037445, + "learning_rate": 0.0003017460237387491, + "loss": 1.8118, + "step": 4893 + }, + { + "epoch": 0.4779296875, + "grad_norm": 0.20497925579547882, + "learning_rate": 0.00030167541410472127, + "loss": 1.8236, + "step": 4894 + }, + { + "epoch": 0.47802734375, + "grad_norm": 0.2557959258556366, + "learning_rate": 0.0003016048018060559, + "loss": 1.828, + "step": 4895 + }, + { + "epoch": 0.478125, + "grad_norm": 0.18973244726657867, + "learning_rate": 0.0003015341868498066, + "loss": 1.8348, + "step": 4896 + }, + { + "epoch": 0.47822265625, + "grad_norm": 0.21844008564949036, + "learning_rate": 0.0003014635692430271, + "loss": 1.8442, + "step": 4897 + }, + { + "epoch": 0.4783203125, + "grad_norm": 0.2102670669555664, + "learning_rate": 0.0003013929489927714, + "loss": 1.8281, + "step": 4898 + }, + { + "epoch": 0.47841796875, + "grad_norm": 0.21705353260040283, + "learning_rate": 0.000301322326106094, + "loss": 1.8495, + "step": 4899 + }, + { + "epoch": 0.478515625, + "grad_norm": 0.20494355261325836, + "learning_rate": 0.0003012517005900494, + "loss": 1.8531, + "step": 4900 + }, + { + "epoch": 0.47861328125, + "grad_norm": 0.24969077110290527, + "learning_rate": 0.0003011810724516924, + "loss": 1.8269, + "step": 4901 + }, + { + "epoch": 0.4787109375, + "grad_norm": 0.18864525854587555, + "learning_rate": 0.00030111044169807834, + "loss": 1.8504, + "step": 4902 + }, + { + "epoch": 0.47880859375, + "grad_norm": 0.2562215328216553, + "learning_rate": 0.00030103980833626236, + "loss": 1.7692, + "step": 4903 + }, + { + "epoch": 0.47890625, + "grad_norm": 0.16926217079162598, + "learning_rate": 0.00030096917237330026, + "loss": 1.8161, + "step": 4904 + }, + { + "epoch": 0.47900390625, + "grad_norm": 0.26131317019462585, + "learning_rate": 0.00030089853381624795, + "loss": 1.8637, + "step": 4905 + }, + { + "epoch": 0.4791015625, + "grad_norm": 0.1805618852376938, + "learning_rate": 0.0003008278926721615, + "loss": 1.8404, + "step": 4906 + }, + { + "epoch": 0.47919921875, + "grad_norm": 0.21555235981941223, + "learning_rate": 0.00030075724894809735, + "loss": 1.8624, + "step": 4907 + }, + { + "epoch": 0.479296875, + "grad_norm": 0.20038826763629913, + "learning_rate": 0.00030068660265111227, + "loss": 1.81, + "step": 4908 + }, + { + "epoch": 0.47939453125, + "grad_norm": 0.20453526079654694, + "learning_rate": 0.0003006159537882631, + "loss": 1.8658, + "step": 4909 + }, + { + "epoch": 0.4794921875, + "grad_norm": 0.23110945522785187, + "learning_rate": 0.0003005453023666072, + "loss": 1.8521, + "step": 4910 + }, + { + "epoch": 0.47958984375, + "grad_norm": 0.25841307640075684, + "learning_rate": 0.0003004746483932018, + "loss": 1.825, + "step": 4911 + }, + { + "epoch": 0.4796875, + "grad_norm": 0.21263115108013153, + "learning_rate": 0.0003004039918751047, + "loss": 1.8036, + "step": 4912 + }, + { + "epoch": 0.47978515625, + "grad_norm": 0.21389488875865936, + "learning_rate": 0.00030033333281937396, + "loss": 1.7746, + "step": 4913 + }, + { + "epoch": 0.4798828125, + "grad_norm": 0.22237850725650787, + "learning_rate": 0.00030026267123306764, + "loss": 1.8436, + "step": 4914 + }, + { + "epoch": 0.47998046875, + "grad_norm": 0.2137548178434372, + "learning_rate": 0.0003001920071232443, + "loss": 1.8037, + "step": 4915 + }, + { + "epoch": 0.480078125, + "grad_norm": 0.1943012923002243, + "learning_rate": 0.0003001213404969626, + "loss": 1.8613, + "step": 4916 + }, + { + "epoch": 0.48017578125, + "grad_norm": 0.1761748194694519, + "learning_rate": 0.00030005067136128157, + "loss": 1.8841, + "step": 4917 + }, + { + "epoch": 0.4802734375, + "grad_norm": 0.20015399158000946, + "learning_rate": 0.00029997999972326036, + "loss": 1.8309, + "step": 4918 + }, + { + "epoch": 0.48037109375, + "grad_norm": 0.19679149985313416, + "learning_rate": 0.00029990932558995854, + "loss": 1.8015, + "step": 4919 + }, + { + "epoch": 0.48046875, + "grad_norm": 0.22215259075164795, + "learning_rate": 0.00029983864896843576, + "loss": 1.837, + "step": 4920 + }, + { + "epoch": 0.48056640625, + "grad_norm": 0.20847636461257935, + "learning_rate": 0.00029976796986575194, + "loss": 1.7845, + "step": 4921 + }, + { + "epoch": 0.4806640625, + "grad_norm": 0.21888545155525208, + "learning_rate": 0.0002996972882889674, + "loss": 1.7938, + "step": 4922 + }, + { + "epoch": 0.48076171875, + "grad_norm": 0.1888756901025772, + "learning_rate": 0.00029962660424514247, + "loss": 1.8435, + "step": 4923 + }, + { + "epoch": 0.480859375, + "grad_norm": 0.2355894297361374, + "learning_rate": 0.0002995559177413381, + "loss": 1.8505, + "step": 4924 + }, + { + "epoch": 0.48095703125, + "grad_norm": 0.18631896376609802, + "learning_rate": 0.00029948522878461495, + "loss": 1.84, + "step": 4925 + }, + { + "epoch": 0.4810546875, + "grad_norm": 0.20177164673805237, + "learning_rate": 0.0002994145373820345, + "loss": 1.8834, + "step": 4926 + }, + { + "epoch": 0.48115234375, + "grad_norm": 0.19149263203144073, + "learning_rate": 0.000299343843540658, + "loss": 1.825, + "step": 4927 + }, + { + "epoch": 0.48125, + "grad_norm": 0.2072160691022873, + "learning_rate": 0.0002992731472675472, + "loss": 1.8557, + "step": 4928 + }, + { + "epoch": 0.48134765625, + "grad_norm": 0.19966818392276764, + "learning_rate": 0.00029920244856976395, + "loss": 1.8274, + "step": 4929 + }, + { + "epoch": 0.4814453125, + "grad_norm": 0.21067434549331665, + "learning_rate": 0.00029913174745437054, + "loss": 1.8083, + "step": 4930 + }, + { + "epoch": 0.48154296875, + "grad_norm": 0.21291854977607727, + "learning_rate": 0.00029906104392842946, + "loss": 1.8408, + "step": 4931 + }, + { + "epoch": 0.481640625, + "grad_norm": 0.24560315907001495, + "learning_rate": 0.0002989903379990031, + "loss": 1.8039, + "step": 4932 + }, + { + "epoch": 0.48173828125, + "grad_norm": 0.21008658409118652, + "learning_rate": 0.0002989196296731547, + "loss": 1.8344, + "step": 4933 + }, + { + "epoch": 0.4818359375, + "grad_norm": 0.1998874843120575, + "learning_rate": 0.00029884891895794706, + "loss": 1.836, + "step": 4934 + }, + { + "epoch": 0.48193359375, + "grad_norm": 0.1839420348405838, + "learning_rate": 0.0002987782058604437, + "loss": 1.8072, + "step": 4935 + }, + { + "epoch": 0.48203125, + "grad_norm": 0.18092377483844757, + "learning_rate": 0.00029870749038770825, + "loss": 1.8041, + "step": 4936 + }, + { + "epoch": 0.48212890625, + "grad_norm": 0.19177129864692688, + "learning_rate": 0.0002986367725468045, + "loss": 1.7981, + "step": 4937 + }, + { + "epoch": 0.4822265625, + "grad_norm": 0.16073240339756012, + "learning_rate": 0.0002985660523447967, + "loss": 1.82, + "step": 4938 + }, + { + "epoch": 0.48232421875, + "grad_norm": 0.19713576138019562, + "learning_rate": 0.0002984953297887489, + "loss": 1.8136, + "step": 4939 + }, + { + "epoch": 0.482421875, + "grad_norm": 0.16476258635520935, + "learning_rate": 0.0002984246048857259, + "loss": 1.853, + "step": 4940 + }, + { + "epoch": 0.48251953125, + "grad_norm": 0.20114773511886597, + "learning_rate": 0.0002983538776427923, + "loss": 1.8006, + "step": 4941 + }, + { + "epoch": 0.4826171875, + "grad_norm": 0.18631845712661743, + "learning_rate": 0.00029828314806701334, + "loss": 1.8018, + "step": 4942 + }, + { + "epoch": 0.48271484375, + "grad_norm": 0.1939045935869217, + "learning_rate": 0.0002982124161654541, + "loss": 1.7903, + "step": 4943 + }, + { + "epoch": 0.4828125, + "grad_norm": 0.22019857168197632, + "learning_rate": 0.0002981416819451801, + "loss": 1.8579, + "step": 4944 + }, + { + "epoch": 0.48291015625, + "grad_norm": 0.25916898250579834, + "learning_rate": 0.0002980709454132571, + "loss": 1.8313, + "step": 4945 + }, + { + "epoch": 0.4830078125, + "grad_norm": 0.19881433248519897, + "learning_rate": 0.00029800020657675107, + "loss": 1.8632, + "step": 4946 + }, + { + "epoch": 0.48310546875, + "grad_norm": 0.23610596358776093, + "learning_rate": 0.0002979294654427282, + "loss": 1.8515, + "step": 4947 + }, + { + "epoch": 0.483203125, + "grad_norm": 0.290463387966156, + "learning_rate": 0.00029785872201825483, + "loss": 1.8176, + "step": 4948 + }, + { + "epoch": 0.48330078125, + "grad_norm": 0.18031997978687286, + "learning_rate": 0.00029778797631039764, + "loss": 1.8665, + "step": 4949 + }, + { + "epoch": 0.4833984375, + "grad_norm": 0.2531370520591736, + "learning_rate": 0.00029771722832622357, + "loss": 1.835, + "step": 4950 + }, + { + "epoch": 0.48349609375, + "grad_norm": 0.22413069009780884, + "learning_rate": 0.0002976464780727995, + "loss": 1.7792, + "step": 4951 + }, + { + "epoch": 0.48359375, + "grad_norm": 0.20282338559627533, + "learning_rate": 0.000297575725557193, + "loss": 1.8828, + "step": 4952 + }, + { + "epoch": 0.48369140625, + "grad_norm": 0.22885966300964355, + "learning_rate": 0.0002975049707864715, + "loss": 1.8764, + "step": 4953 + }, + { + "epoch": 0.4837890625, + "grad_norm": 0.23277723789215088, + "learning_rate": 0.0002974342137677028, + "loss": 1.8568, + "step": 4954 + }, + { + "epoch": 0.48388671875, + "grad_norm": 0.18464137613773346, + "learning_rate": 0.00029736345450795485, + "loss": 1.8235, + "step": 4955 + }, + { + "epoch": 0.483984375, + "grad_norm": 0.2433868944644928, + "learning_rate": 0.0002972926930142959, + "loss": 1.8362, + "step": 4956 + }, + { + "epoch": 0.48408203125, + "grad_norm": 0.2033810317516327, + "learning_rate": 0.00029722192929379455, + "loss": 1.8419, + "step": 4957 + }, + { + "epoch": 0.4841796875, + "grad_norm": 0.19732214510440826, + "learning_rate": 0.0002971511633535193, + "loss": 1.8475, + "step": 4958 + }, + { + "epoch": 0.48427734375, + "grad_norm": 0.2697364091873169, + "learning_rate": 0.000297080395200539, + "loss": 1.8306, + "step": 4959 + }, + { + "epoch": 0.484375, + "grad_norm": 0.2421739399433136, + "learning_rate": 0.00029700962484192277, + "loss": 1.8013, + "step": 4960 + }, + { + "epoch": 0.48447265625, + "grad_norm": 0.1866464763879776, + "learning_rate": 0.00029693885228474013, + "loss": 1.8413, + "step": 4961 + }, + { + "epoch": 0.4845703125, + "grad_norm": 0.23989856243133545, + "learning_rate": 0.00029686807753606046, + "loss": 1.8138, + "step": 4962 + }, + { + "epoch": 0.48466796875, + "grad_norm": 0.1735045164823532, + "learning_rate": 0.00029679730060295356, + "loss": 1.8457, + "step": 4963 + }, + { + "epoch": 0.484765625, + "grad_norm": 0.2334936559200287, + "learning_rate": 0.0002967265214924895, + "loss": 1.8282, + "step": 4964 + }, + { + "epoch": 0.48486328125, + "grad_norm": 0.19146543741226196, + "learning_rate": 0.0002966557402117384, + "loss": 1.844, + "step": 4965 + }, + { + "epoch": 0.4849609375, + "grad_norm": 0.20807620882987976, + "learning_rate": 0.00029658495676777077, + "loss": 1.8132, + "step": 4966 + }, + { + "epoch": 0.48505859375, + "grad_norm": 0.2114933729171753, + "learning_rate": 0.0002965141711676569, + "loss": 1.8179, + "step": 4967 + }, + { + "epoch": 0.48515625, + "grad_norm": 0.20859596133232117, + "learning_rate": 0.0002964433834184683, + "loss": 1.8603, + "step": 4968 + }, + { + "epoch": 0.48525390625, + "grad_norm": 0.1918727159500122, + "learning_rate": 0.0002963725935272755, + "loss": 1.7765, + "step": 4969 + }, + { + "epoch": 0.4853515625, + "grad_norm": 0.21384978294372559, + "learning_rate": 0.00029630180150114985, + "loss": 1.8126, + "step": 4970 + }, + { + "epoch": 0.48544921875, + "grad_norm": 0.2083677351474762, + "learning_rate": 0.0002962310073471631, + "loss": 1.8557, + "step": 4971 + }, + { + "epoch": 0.485546875, + "grad_norm": 0.1978531926870346, + "learning_rate": 0.0002961602110723868, + "loss": 1.8052, + "step": 4972 + }, + { + "epoch": 0.48564453125, + "grad_norm": 0.27248382568359375, + "learning_rate": 0.00029608941268389284, + "loss": 1.8608, + "step": 4973 + }, + { + "epoch": 0.4857421875, + "grad_norm": 0.19011075794696808, + "learning_rate": 0.0002960186121887533, + "loss": 1.8241, + "step": 4974 + }, + { + "epoch": 0.48583984375, + "grad_norm": 0.27825161814689636, + "learning_rate": 0.00029594780959404075, + "loss": 1.825, + "step": 4975 + }, + { + "epoch": 0.4859375, + "grad_norm": 0.23274178802967072, + "learning_rate": 0.00029587700490682755, + "loss": 1.8529, + "step": 4976 + }, + { + "epoch": 0.48603515625, + "grad_norm": 0.22092591226100922, + "learning_rate": 0.00029580619813418643, + "loss": 1.8793, + "step": 4977 + }, + { + "epoch": 0.4861328125, + "grad_norm": 0.24535295367240906, + "learning_rate": 0.00029573538928319063, + "loss": 1.8331, + "step": 4978 + }, + { + "epoch": 0.48623046875, + "grad_norm": 0.2540774941444397, + "learning_rate": 0.000295664578360913, + "loss": 1.8456, + "step": 4979 + }, + { + "epoch": 0.486328125, + "grad_norm": 0.25606876611709595, + "learning_rate": 0.0002955937653744271, + "loss": 1.796, + "step": 4980 + }, + { + "epoch": 0.48642578125, + "grad_norm": 0.23073682188987732, + "learning_rate": 0.0002955229503308064, + "loss": 1.8139, + "step": 4981 + }, + { + "epoch": 0.4865234375, + "grad_norm": 0.2704898416996002, + "learning_rate": 0.00029545213323712487, + "loss": 1.857, + "step": 4982 + }, + { + "epoch": 0.48662109375, + "grad_norm": 0.18947190046310425, + "learning_rate": 0.0002953813141004564, + "loss": 1.8144, + "step": 4983 + }, + { + "epoch": 0.48671875, + "grad_norm": 0.25032109022140503, + "learning_rate": 0.00029531049292787524, + "loss": 1.8256, + "step": 4984 + }, + { + "epoch": 0.48681640625, + "grad_norm": 0.2130419909954071, + "learning_rate": 0.00029523966972645575, + "loss": 1.8148, + "step": 4985 + }, + { + "epoch": 0.4869140625, + "grad_norm": 0.19153909385204315, + "learning_rate": 0.0002951688445032726, + "loss": 1.796, + "step": 4986 + }, + { + "epoch": 0.48701171875, + "grad_norm": 0.2378961592912674, + "learning_rate": 0.0002950980172654005, + "loss": 1.8424, + "step": 4987 + }, + { + "epoch": 0.487109375, + "grad_norm": 0.21440565586090088, + "learning_rate": 0.00029502718801991453, + "loss": 1.8671, + "step": 4988 + }, + { + "epoch": 0.48720703125, + "grad_norm": 0.22810520231723785, + "learning_rate": 0.0002949563567738899, + "loss": 1.8683, + "step": 4989 + }, + { + "epoch": 0.4873046875, + "grad_norm": 0.19989818334579468, + "learning_rate": 0.0002948855235344021, + "loss": 1.8653, + "step": 4990 + }, + { + "epoch": 0.48740234375, + "grad_norm": 0.1847819685935974, + "learning_rate": 0.0002948146883085266, + "loss": 1.8064, + "step": 4991 + }, + { + "epoch": 0.4875, + "grad_norm": 0.201404869556427, + "learning_rate": 0.0002947438511033393, + "loss": 1.8708, + "step": 4992 + }, + { + "epoch": 0.48759765625, + "grad_norm": 0.18044066429138184, + "learning_rate": 0.0002946730119259162, + "loss": 1.8017, + "step": 4993 + }, + { + "epoch": 0.4876953125, + "grad_norm": 0.19026845693588257, + "learning_rate": 0.00029460217078333353, + "loss": 1.8469, + "step": 4994 + }, + { + "epoch": 0.48779296875, + "grad_norm": 0.18897053599357605, + "learning_rate": 0.0002945313276826675, + "loss": 1.8525, + "step": 4995 + }, + { + "epoch": 0.487890625, + "grad_norm": 0.18522904813289642, + "learning_rate": 0.0002944604826309951, + "loss": 1.8473, + "step": 4996 + }, + { + "epoch": 0.48798828125, + "grad_norm": 0.20513242483139038, + "learning_rate": 0.0002943896356353927, + "loss": 1.8781, + "step": 4997 + }, + { + "epoch": 0.4880859375, + "grad_norm": 0.19440455734729767, + "learning_rate": 0.0002943187867029375, + "loss": 1.8408, + "step": 4998 + }, + { + "epoch": 0.48818359375, + "grad_norm": 0.18897245824337006, + "learning_rate": 0.0002942479358407068, + "loss": 1.8311, + "step": 4999 + }, + { + "epoch": 0.48828125, + "grad_norm": 0.18528372049331665, + "learning_rate": 0.0002941770830557777, + "loss": 1.7681, + "step": 5000 + } + ], + "logging_steps": 1.0, + "max_steps": 10240, + "num_input_tokens_seen": 0, + "num_train_epochs": 9223372036854775807, + "save_steps": 200, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": false + }, + "attributes": {} + } + }, + "total_flos": 3.061252972609536e+19, + "train_batch_size": 16, + "trial_name": null, + "trial_params": null +}