| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9254621347644604, | |
| "eval_steps": 50, | |
| "global_step": 97, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.009540846750149075, | |
| "grad_norm": 0.7367311120033264, | |
| "learning_rate": 2e-05, | |
| "loss": 3.3667, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.009540846750149075, | |
| "eval_loss": 3.5421202182769775, | |
| "eval_runtime": 46.8738, | |
| "eval_samples_per_second": 7.552, | |
| "eval_steps_per_second": 1.899, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01908169350029815, | |
| "grad_norm": 0.9371811151504517, | |
| "learning_rate": 4e-05, | |
| "loss": 2.9695, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.028622540250447227, | |
| "grad_norm": 0.614778459072113, | |
| "learning_rate": 6e-05, | |
| "loss": 3.2482, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0381633870005963, | |
| "grad_norm": 0.7607264518737793, | |
| "learning_rate": 8e-05, | |
| "loss": 3.3045, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.04770423375074538, | |
| "grad_norm": 0.792961597442627, | |
| "learning_rate": 0.0001, | |
| "loss": 3.299, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.057245080500894455, | |
| "grad_norm": 0.9246425032615662, | |
| "learning_rate": 0.00012, | |
| "loss": 3.2984, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.06678592725104353, | |
| "grad_norm": 0.9336756467819214, | |
| "learning_rate": 0.00014, | |
| "loss": 3.0915, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0763267740011926, | |
| "grad_norm": 1.0689278841018677, | |
| "learning_rate": 0.00016, | |
| "loss": 3.1456, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.08586762075134168, | |
| "grad_norm": 0.6929656267166138, | |
| "learning_rate": 0.00018, | |
| "loss": 2.8801, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.09540846750149076, | |
| "grad_norm": 0.9856171011924744, | |
| "learning_rate": 0.0002, | |
| "loss": 2.6816, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.10494931425163984, | |
| "grad_norm": 0.9085230827331543, | |
| "learning_rate": 0.0001999348095389677, | |
| "loss": 2.6598, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.11449016100178891, | |
| "grad_norm": 0.8404436707496643, | |
| "learning_rate": 0.000199739323151795, | |
| "loss": 2.6531, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.12403100775193798, | |
| "grad_norm": 0.8586527109146118, | |
| "learning_rate": 0.00019941379571543596, | |
| "loss": 2.6339, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.13357185450208706, | |
| "grad_norm": 0.851191520690918, | |
| "learning_rate": 0.00019895865165556377, | |
| "loss": 2.5046, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.14311270125223613, | |
| "grad_norm": 0.6186274886131287, | |
| "learning_rate": 0.00019837448439320027, | |
| "loss": 2.357, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.1526535480023852, | |
| "grad_norm": 0.6838149428367615, | |
| "learning_rate": 0.00019766205557100868, | |
| "loss": 2.2243, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.16219439475253428, | |
| "grad_norm": 1.2345480918884277, | |
| "learning_rate": 0.00019682229406025635, | |
| "loss": 2.209, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.17173524150268335, | |
| "grad_norm": 0.5735088586807251, | |
| "learning_rate": 0.00019585629474974415, | |
| "loss": 2.5145, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.18127608825283245, | |
| "grad_norm": 0.551199734210968, | |
| "learning_rate": 0.00019476531711828027, | |
| "loss": 2.4064, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.19081693500298152, | |
| "grad_norm": 0.6080546379089355, | |
| "learning_rate": 0.0001935507835925601, | |
| "loss": 2.5871, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2003577817531306, | |
| "grad_norm": 0.6316207051277161, | |
| "learning_rate": 0.00019221427769259333, | |
| "loss": 2.4882, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.20989862850327967, | |
| "grad_norm": 0.5335537195205688, | |
| "learning_rate": 0.00019075754196709572, | |
| "loss": 2.3722, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.21943947525342875, | |
| "grad_norm": 0.516207754611969, | |
| "learning_rate": 0.00018918247572153823, | |
| "loss": 2.4403, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.22898032200357782, | |
| "grad_norm": 0.47695738077163696, | |
| "learning_rate": 0.00018749113254181498, | |
| "loss": 2.2934, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.2385211687537269, | |
| "grad_norm": 0.4737792909145355, | |
| "learning_rate": 0.00018568571761675893, | |
| "loss": 2.1638, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.24806201550387597, | |
| "grad_norm": 0.40663591027259827, | |
| "learning_rate": 0.00018376858486299647, | |
| "loss": 2.0649, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.25760286225402507, | |
| "grad_norm": 0.4352891147136688, | |
| "learning_rate": 0.00018174223385588917, | |
| "loss": 2.2987, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.2671437090041741, | |
| "grad_norm": 0.4194694757461548, | |
| "learning_rate": 0.00017960930657056438, | |
| "loss": 2.272, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.2766845557543232, | |
| "grad_norm": 0.5331905484199524, | |
| "learning_rate": 0.00017737258393728364, | |
| "loss": 2.3588, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.28622540250447226, | |
| "grad_norm": 0.4592475891113281, | |
| "learning_rate": 0.00017503498221564025, | |
| "loss": 2.3871, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.29576624925462136, | |
| "grad_norm": 0.4838412404060364, | |
| "learning_rate": 0.0001725995491923131, | |
| "loss": 2.1406, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.3053070960047704, | |
| "grad_norm": 0.4136384427547455, | |
| "learning_rate": 0.00017006946020733425, | |
| "loss": 2.1532, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.3148479427549195, | |
| "grad_norm": 0.41302838921546936, | |
| "learning_rate": 0.0001674480140140514, | |
| "loss": 2.1363, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.32438878950506855, | |
| "grad_norm": 0.44015756249427795, | |
| "learning_rate": 0.00016473862847818277, | |
| "loss": 2.265, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.33392963625521765, | |
| "grad_norm": 0.44776979088783264, | |
| "learning_rate": 0.0001619448361215723, | |
| "loss": 2.2622, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.3434704830053667, | |
| "grad_norm": 0.4290170967578888, | |
| "learning_rate": 0.0001590702795164551, | |
| "loss": 2.0401, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.3530113297555158, | |
| "grad_norm": 0.48702678084373474, | |
| "learning_rate": 0.00015611870653623825, | |
| "loss": 2.1099, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.3625521765056649, | |
| "grad_norm": 0.5152214765548706, | |
| "learning_rate": 0.0001530939654689887, | |
| "loss": 2.1686, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.37209302325581395, | |
| "grad_norm": 0.4602924585342407, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 2.1263, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.38163387000596305, | |
| "grad_norm": 0.47519952058792114, | |
| "learning_rate": 0.00014684084406997903, | |
| "loss": 1.9896, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3911747167561121, | |
| "grad_norm": 0.5167194604873657, | |
| "learning_rate": 0.00014362061661555675, | |
| "loss": 2.2722, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.4007155635062612, | |
| "grad_norm": 0.48113343119621277, | |
| "learning_rate": 0.00014034351619898088, | |
| "loss": 2.292, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.41025641025641024, | |
| "grad_norm": 0.484389066696167, | |
| "learning_rate": 0.00013701381553399145, | |
| "loss": 2.2551, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.41979725700655934, | |
| "grad_norm": 0.4332115650177002, | |
| "learning_rate": 0.0001336358559150175, | |
| "loss": 1.9839, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.4293381037567084, | |
| "grad_norm": 0.4608907401561737, | |
| "learning_rate": 0.00013021404155695725, | |
| "loss": 1.9983, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.4388789505068575, | |
| "grad_norm": 0.49616098403930664, | |
| "learning_rate": 0.00012675283385292212, | |
| "loss": 2.1903, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.44841979725700654, | |
| "grad_norm": 0.573137640953064, | |
| "learning_rate": 0.00012325674555743106, | |
| "loss": 2.1866, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.45796064400715564, | |
| "grad_norm": 0.45796340703964233, | |
| "learning_rate": 0.00011973033490264001, | |
| "loss": 2.0213, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.46750149075730474, | |
| "grad_norm": 0.5226765871047974, | |
| "learning_rate": 0.0001161781996552765, | |
| "loss": 2.1563, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.4770423375074538, | |
| "grad_norm": 0.5040608644485474, | |
| "learning_rate": 0.00011260497112202895, | |
| "loss": 2.1839, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.4770423375074538, | |
| "eval_loss": 2.176058769226074, | |
| "eval_runtime": 47.4353, | |
| "eval_samples_per_second": 7.463, | |
| "eval_steps_per_second": 1.876, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.4865831842576029, | |
| "grad_norm": 0.4927314519882202, | |
| "learning_rate": 0.00010901530811120655, | |
| "loss": 2.0997, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.49612403100775193, | |
| "grad_norm": 0.5394675731658936, | |
| "learning_rate": 0.00010541389085854176, | |
| "loss": 2.2713, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.505664877757901, | |
| "grad_norm": 0.5610401630401611, | |
| "learning_rate": 0.00010180541492505604, | |
| "loss": 2.1373, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.5152057245080501, | |
| "grad_norm": 0.5036469101905823, | |
| "learning_rate": 9.819458507494394e-05, | |
| "loss": 2.2499, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.5247465712581991, | |
| "grad_norm": 0.46630653738975525, | |
| "learning_rate": 9.458610914145826e-05, | |
| "loss": 1.9732, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.5342874180083482, | |
| "grad_norm": 0.47023531794548035, | |
| "learning_rate": 9.098469188879349e-05, | |
| "loss": 2.1248, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.5438282647584973, | |
| "grad_norm": 0.4970254898071289, | |
| "learning_rate": 8.739502887797107e-05, | |
| "loss": 2.0411, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.5533691115086464, | |
| "grad_norm": 0.5040042400360107, | |
| "learning_rate": 8.382180034472353e-05, | |
| "loss": 2.2428, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.5629099582587954, | |
| "grad_norm": 0.5225110650062561, | |
| "learning_rate": 8.026966509736001e-05, | |
| "loss": 2.1539, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.5724508050089445, | |
| "grad_norm": 0.5296663641929626, | |
| "learning_rate": 7.674325444256899e-05, | |
| "loss": 2.1092, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5819916517590936, | |
| "grad_norm": 0.4598320424556732, | |
| "learning_rate": 7.324716614707793e-05, | |
| "loss": 2.061, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.5915324985092427, | |
| "grad_norm": 0.5481743216514587, | |
| "learning_rate": 6.978595844304271e-05, | |
| "loss": 2.0384, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.6010733452593918, | |
| "grad_norm": 0.569454550743103, | |
| "learning_rate": 6.636414408498249e-05, | |
| "loss": 2.2044, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.6106141920095408, | |
| "grad_norm": 0.5209865570068359, | |
| "learning_rate": 6.298618446600856e-05, | |
| "loss": 2.0685, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.6201550387596899, | |
| "grad_norm": 0.5345067977905273, | |
| "learning_rate": 5.965648380101916e-05, | |
| "loss": 2.1992, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.629695885509839, | |
| "grad_norm": 0.46630385518074036, | |
| "learning_rate": 5.6379383384443255e-05, | |
| "loss": 2.2016, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.6392367322599881, | |
| "grad_norm": 0.5166553854942322, | |
| "learning_rate": 5.3159155930021e-05, | |
| "loss": 2.1986, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.6487775790101371, | |
| "grad_norm": 0.4024442136287689, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 1.9141, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.6583184257602862, | |
| "grad_norm": 0.5243792533874512, | |
| "learning_rate": 4.6906034531011346e-05, | |
| "loss": 2.3524, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.6678592725104353, | |
| "grad_norm": 0.5335293412208557, | |
| "learning_rate": 4.388129346376178e-05, | |
| "loss": 2.1225, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6774001192605844, | |
| "grad_norm": 0.5144345164299011, | |
| "learning_rate": 4.092972048354491e-05, | |
| "loss": 2.1678, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.6869409660107334, | |
| "grad_norm": 0.4751684069633484, | |
| "learning_rate": 3.80551638784277e-05, | |
| "loss": 2.1261, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.6964818127608825, | |
| "grad_norm": 0.5534799098968506, | |
| "learning_rate": 3.5261371521817244e-05, | |
| "loss": 2.3748, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.7060226595110316, | |
| "grad_norm": 0.48840370774269104, | |
| "learning_rate": 3.2551985985948616e-05, | |
| "loss": 2.0223, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.7155635062611807, | |
| "grad_norm": 0.45086348056793213, | |
| "learning_rate": 2.993053979266577e-05, | |
| "loss": 2.0576, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.7251043530113298, | |
| "grad_norm": 0.4344973862171173, | |
| "learning_rate": 2.7400450807686938e-05, | |
| "loss": 1.9448, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.7346451997614788, | |
| "grad_norm": 0.5358251333236694, | |
| "learning_rate": 2.496501778435977e-05, | |
| "loss": 2.0693, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 0.7441860465116279, | |
| "grad_norm": 0.550864577293396, | |
| "learning_rate": 2.2627416062716366e-05, | |
| "loss": 2.2065, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.753726893261777, | |
| "grad_norm": 0.594879150390625, | |
| "learning_rate": 2.0390693429435627e-05, | |
| "loss": 2.2258, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 0.7632677400119261, | |
| "grad_norm": 0.5433897376060486, | |
| "learning_rate": 1.825776614411082e-05, | |
| "loss": 2.0444, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.7728085867620751, | |
| "grad_norm": 0.5564923882484436, | |
| "learning_rate": 1.6231415137003537e-05, | |
| "loss": 2.2308, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.7823494335122242, | |
| "grad_norm": 0.5145489573478699, | |
| "learning_rate": 1.4314282383241096e-05, | |
| "loss": 2.1803, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 0.7918902802623733, | |
| "grad_norm": 0.5326076745986938, | |
| "learning_rate": 1.2508867458185037e-05, | |
| "loss": 2.1118, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 0.8014311270125224, | |
| "grad_norm": 0.5149711966514587, | |
| "learning_rate": 1.0817524278461776e-05, | |
| "loss": 2.2356, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.8109719737626714, | |
| "grad_norm": 0.545981228351593, | |
| "learning_rate": 9.242458032904311e-06, | |
| "loss": 2.0592, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.8205128205128205, | |
| "grad_norm": 0.48461177945137024, | |
| "learning_rate": 7.785722307406684e-06, | |
| "loss": 2.0525, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 0.8300536672629696, | |
| "grad_norm": 0.4771486818790436, | |
| "learning_rate": 6.4492164074399065e-06, | |
| "loss": 2.1279, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.8395945140131187, | |
| "grad_norm": 0.5691129565238953, | |
| "learning_rate": 5.2346828817197655e-06, | |
| "loss": 2.2043, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.8491353607632678, | |
| "grad_norm": 0.5064877867698669, | |
| "learning_rate": 4.143705250255869e-06, | |
| "loss": 2.1349, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 0.8586762075134168, | |
| "grad_norm": 0.4522230923175812, | |
| "learning_rate": 3.1777059397436692e-06, | |
| "loss": 2.177, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.8682170542635659, | |
| "grad_norm": 0.6604287028312683, | |
| "learning_rate": 2.3379444289913342e-06, | |
| "loss": 2.0478, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 0.877757901013715, | |
| "grad_norm": 0.4626028835773468, | |
| "learning_rate": 1.6255156067997323e-06, | |
| "loss": 2.331, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.8872987477638641, | |
| "grad_norm": 0.5617081522941589, | |
| "learning_rate": 1.0413483444362771e-06, | |
| "loss": 2.0815, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.8968395945140131, | |
| "grad_norm": 0.48092931509017944, | |
| "learning_rate": 5.862042845640403e-07, | |
| "loss": 1.9864, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 0.9063804412641622, | |
| "grad_norm": 0.4518766701221466, | |
| "learning_rate": 2.606768482050215e-07, | |
| "loss": 2.0202, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.9159212880143113, | |
| "grad_norm": 0.5857305526733398, | |
| "learning_rate": 6.519046103230508e-08, | |
| "loss": 2.1958, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.9254621347644604, | |
| "grad_norm": 0.5706104636192322, | |
| "learning_rate": 0.0, | |
| "loss": 2.1415, | |
| "step": 97 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 97, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 2, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.1683682015195955e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |