| { | |
| "best_metric": 1.0, | |
| "best_model_checkpoint": "test_long_tokens/checkpoint-70", | |
| "epoch": 14.285714285714286, | |
| "eval_steps": 500, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14285714285714285, | |
| "grad_norm": 18.59529685974121, | |
| "learning_rate": 5e-06, | |
| "loss": 1.637, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 20.817951202392578, | |
| "learning_rate": 1e-05, | |
| "loss": 1.6372, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.42857142857142855, | |
| "grad_norm": 19.524600982666016, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.4306, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 15.0996675491333, | |
| "learning_rate": 2e-05, | |
| "loss": 1.2657, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.7142857142857143, | |
| "grad_norm": 12.507349967956543, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.8948, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.8571428571428571, | |
| "grad_norm": 8.167460441589355, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5404, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 5.33146858215332, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.5374, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8645276292335116, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.7161440253257751, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 0.2378, | |
| "eval_samples_per_second": 12.618, | |
| "eval_steps_per_second": 12.618, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 2.557964324951172, | |
| "learning_rate": 4e-05, | |
| "loss": 0.2147, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.2857142857142856, | |
| "grad_norm": 4.4479289054870605, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5833, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 5.119785308837891, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6062, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.5714285714285714, | |
| "grad_norm": 3.064371347427368, | |
| "learning_rate": 4.9444444444444446e-05, | |
| "loss": 0.451, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 1.8660037517547607, | |
| "learning_rate": 4.888888888888889e-05, | |
| "loss": 0.3188, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.8571428571428572, | |
| "grad_norm": 2.083087205886841, | |
| "learning_rate": 4.8333333333333334e-05, | |
| "loss": 0.2287, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.0547335147857666, | |
| "learning_rate": 4.7777777777777784e-05, | |
| "loss": 0.3423, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8645276292335116, | |
| "eval_f1": 0.0, | |
| "eval_loss": 0.459845632314682, | |
| "eval_precision": 0.0, | |
| "eval_recall": 0.0, | |
| "eval_runtime": 0.2226, | |
| "eval_samples_per_second": 13.474, | |
| "eval_steps_per_second": 13.474, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.142857142857143, | |
| "grad_norm": 2.206693410873413, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.2449, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 0.9722664952278137, | |
| "learning_rate": 4.666666666666667e-05, | |
| "loss": 0.1875, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.4285714285714284, | |
| "grad_norm": 1.500380039215088, | |
| "learning_rate": 4.6111111111111115e-05, | |
| "loss": 0.1429, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.571428571428571, | |
| "grad_norm": 1.9801063537597656, | |
| "learning_rate": 4.555555555555556e-05, | |
| "loss": 0.1752, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.7142857142857144, | |
| "grad_norm": 2.65030837059021, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.0763, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 3.4686992168426514, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.168, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 6.0364298820495605, | |
| "learning_rate": 4.388888888888889e-05, | |
| "loss": 0.1813, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.966131907308378, | |
| "eval_f1": 0.9019607843137256, | |
| "eval_loss": 0.14405852556228638, | |
| "eval_precision": 0.8846153846153846, | |
| "eval_recall": 0.92, | |
| "eval_runtime": 0.2168, | |
| "eval_samples_per_second": 13.839, | |
| "eval_steps_per_second": 13.839, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 3.142857142857143, | |
| "grad_norm": 0.7663143873214722, | |
| "learning_rate": 4.3333333333333334e-05, | |
| "loss": 0.0552, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 3.2857142857142856, | |
| "grad_norm": 1.2191542387008667, | |
| "learning_rate": 4.277777777777778e-05, | |
| "loss": 0.1005, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 3.4285714285714284, | |
| "grad_norm": 7.446482181549072, | |
| "learning_rate": 4.222222222222222e-05, | |
| "loss": 0.0559, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 3.571428571428571, | |
| "grad_norm": 1.18604576587677, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.0494, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.7142857142857144, | |
| "grad_norm": 0.6873874664306641, | |
| "learning_rate": 4.111111111111111e-05, | |
| "loss": 0.0533, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 3.857142857142857, | |
| "grad_norm": 2.966907024383545, | |
| "learning_rate": 4.055555555555556e-05, | |
| "loss": 0.0625, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.066542863845825, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0734, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9732620320855615, | |
| "eval_f1": 0.8000000000000002, | |
| "eval_loss": 0.13226450979709625, | |
| "eval_precision": 0.8, | |
| "eval_recall": 0.8, | |
| "eval_runtime": 0.2294, | |
| "eval_samples_per_second": 13.077, | |
| "eval_steps_per_second": 13.077, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 4.142857142857143, | |
| "grad_norm": 0.8713329434394836, | |
| "learning_rate": 3.944444444444445e-05, | |
| "loss": 0.0673, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 0.306024968624115, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.0146, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.428571428571429, | |
| "grad_norm": 0.3102547526359558, | |
| "learning_rate": 3.8333333333333334e-05, | |
| "loss": 0.0169, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "grad_norm": 5.933145046234131, | |
| "learning_rate": 3.777777777777778e-05, | |
| "loss": 0.0753, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 4.714285714285714, | |
| "grad_norm": 1.7290794849395752, | |
| "learning_rate": 3.722222222222222e-05, | |
| "loss": 0.0668, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 4.857142857142857, | |
| "grad_norm": 8.295025825500488, | |
| "learning_rate": 3.6666666666666666e-05, | |
| "loss": 0.0786, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.47509345412254333, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.0124, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9928698752228164, | |
| "eval_f1": 0.8235294117647058, | |
| "eval_loss": 0.0685623362660408, | |
| "eval_precision": 0.8076923076923077, | |
| "eval_recall": 0.84, | |
| "eval_runtime": 0.2188, | |
| "eval_samples_per_second": 13.709, | |
| "eval_steps_per_second": 13.709, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 5.142857142857143, | |
| "grad_norm": 1.3178348541259766, | |
| "learning_rate": 3.555555555555556e-05, | |
| "loss": 0.0303, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 5.285714285714286, | |
| "grad_norm": 2.2010915279388428, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.045, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 5.428571428571429, | |
| "grad_norm": 1.1045790910720825, | |
| "learning_rate": 3.444444444444445e-05, | |
| "loss": 0.0145, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 5.571428571428571, | |
| "grad_norm": 1.0724176168441772, | |
| "learning_rate": 3.388888888888889e-05, | |
| "loss": 0.0063, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 0.7634330987930298, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.0176, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.857142857142857, | |
| "grad_norm": 0.48025190830230713, | |
| "learning_rate": 3.277777777777778e-05, | |
| "loss": 0.0228, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.21735815703868866, | |
| "learning_rate": 3.222222222222223e-05, | |
| "loss": 0.0034, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.983957219251337, | |
| "eval_f1": 0.7058823529411765, | |
| "eval_loss": 0.16237980127334595, | |
| "eval_precision": 0.6923076923076923, | |
| "eval_recall": 0.72, | |
| "eval_runtime": 0.2281, | |
| "eval_samples_per_second": 13.15, | |
| "eval_steps_per_second": 13.15, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 6.142857142857143, | |
| "grad_norm": 1.1637978553771973, | |
| "learning_rate": 3.1666666666666666e-05, | |
| "loss": 0.0137, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 6.285714285714286, | |
| "grad_norm": 0.21785947680473328, | |
| "learning_rate": 3.111111111111111e-05, | |
| "loss": 0.0039, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 6.428571428571429, | |
| "grad_norm": 20.79886245727539, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.1387, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 6.571428571428571, | |
| "grad_norm": 7.047394752502441, | |
| "learning_rate": 3e-05, | |
| "loss": 0.019, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 6.714285714285714, | |
| "grad_norm": 0.3814934194087982, | |
| "learning_rate": 2.9444444444444448e-05, | |
| "loss": 0.0059, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 6.857142857142857, | |
| "grad_norm": 0.4642070233821869, | |
| "learning_rate": 2.8888888888888888e-05, | |
| "loss": 0.005, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.06280206888914108, | |
| "learning_rate": 2.8333333333333335e-05, | |
| "loss": 0.0023, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9946524064171123, | |
| "eval_f1": 0.830188679245283, | |
| "eval_loss": 0.018056461587548256, | |
| "eval_precision": 0.7857142857142857, | |
| "eval_recall": 0.88, | |
| "eval_runtime": 0.2418, | |
| "eval_samples_per_second": 12.409, | |
| "eval_steps_per_second": 12.409, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 7.142857142857143, | |
| "grad_norm": 0.042585261166095734, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.002, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 7.285714285714286, | |
| "grad_norm": 3.1652538776397705, | |
| "learning_rate": 2.7222222222222223e-05, | |
| "loss": 0.0085, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 7.428571428571429, | |
| "grad_norm": 0.997791051864624, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 0.0154, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 7.571428571428571, | |
| "grad_norm": 0.2884855568408966, | |
| "learning_rate": 2.6111111111111114e-05, | |
| "loss": 0.0021, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 7.714285714285714, | |
| "grad_norm": 0.39060717821121216, | |
| "learning_rate": 2.5555555555555554e-05, | |
| "loss": 0.0026, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 7.857142857142857, | |
| "grad_norm": 15.578678131103516, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.031, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.1004723384976387, | |
| "learning_rate": 2.4444444444444445e-05, | |
| "loss": 0.0023, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9946524064171123, | |
| "eval_f1": 0.8627450980392156, | |
| "eval_loss": 0.054031386971473694, | |
| "eval_precision": 0.8461538461538461, | |
| "eval_recall": 0.88, | |
| "eval_runtime": 0.3115, | |
| "eval_samples_per_second": 9.631, | |
| "eval_steps_per_second": 9.631, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 8.142857142857142, | |
| "grad_norm": 0.5730704665184021, | |
| "learning_rate": 2.3888888888888892e-05, | |
| "loss": 0.005, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 8.285714285714286, | |
| "grad_norm": 0.08043497800827026, | |
| "learning_rate": 2.3333333333333336e-05, | |
| "loss": 0.0023, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 8.428571428571429, | |
| "grad_norm": 0.03008083440363407, | |
| "learning_rate": 2.277777777777778e-05, | |
| "loss": 0.0012, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "grad_norm": 0.03248002007603645, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.0008, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 8.714285714285714, | |
| "grad_norm": 0.04838703200221062, | |
| "learning_rate": 2.1666666666666667e-05, | |
| "loss": 0.0013, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 8.857142857142858, | |
| "grad_norm": 0.33141571283340454, | |
| "learning_rate": 2.111111111111111e-05, | |
| "loss": 0.0022, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 1.7117937803268433, | |
| "learning_rate": 2.0555555555555555e-05, | |
| "loss": 0.0069, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9269162210338681, | |
| "eval_f1": 0.5, | |
| "eval_loss": 0.5584124326705933, | |
| "eval_precision": 0.5789473684210527, | |
| "eval_recall": 0.44, | |
| "eval_runtime": 0.2368, | |
| "eval_samples_per_second": 12.67, | |
| "eval_steps_per_second": 12.67, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 9.142857142857142, | |
| "grad_norm": 0.3541106581687927, | |
| "learning_rate": 2e-05, | |
| "loss": 0.003, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 9.285714285714286, | |
| "grad_norm": 6.2997145652771, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.0145, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 9.428571428571429, | |
| "grad_norm": 3.1364083290100098, | |
| "learning_rate": 1.888888888888889e-05, | |
| "loss": 0.0043, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 9.571428571428571, | |
| "grad_norm": 0.018303165212273598, | |
| "learning_rate": 1.8333333333333333e-05, | |
| "loss": 0.0007, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 9.714285714285714, | |
| "grad_norm": 0.06226564571261406, | |
| "learning_rate": 1.777777777777778e-05, | |
| "loss": 0.0011, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 9.857142857142858, | |
| "grad_norm": 0.18134748935699463, | |
| "learning_rate": 1.7222222222222224e-05, | |
| "loss": 0.0014, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.013242754153907299, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.0006, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 1.0, | |
| "eval_f1": 1.0, | |
| "eval_loss": 0.004223768133670092, | |
| "eval_precision": 1.0, | |
| "eval_recall": 1.0, | |
| "eval_runtime": 0.2664, | |
| "eval_samples_per_second": 11.263, | |
| "eval_steps_per_second": 11.263, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 10.142857142857142, | |
| "grad_norm": 0.5124610662460327, | |
| "learning_rate": 1.6111111111111115e-05, | |
| "loss": 0.0017, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 10.285714285714286, | |
| "grad_norm": 0.014037979766726494, | |
| "learning_rate": 1.5555555555555555e-05, | |
| "loss": 0.0008, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 10.428571428571429, | |
| "grad_norm": 0.020040445029735565, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.001, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 10.571428571428571, | |
| "grad_norm": 0.0177422147244215, | |
| "learning_rate": 1.4444444444444444e-05, | |
| "loss": 0.0007, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 10.714285714285714, | |
| "grad_norm": 0.02175714634358883, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.0011, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 10.857142857142858, | |
| "grad_norm": 0.016904808580875397, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.0009, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.7300955653190613, | |
| "learning_rate": 1.2777777777777777e-05, | |
| "loss": 0.0026, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 1.0, | |
| "eval_f1": 1.0, | |
| "eval_loss": 0.0016724281013011932, | |
| "eval_precision": 1.0, | |
| "eval_recall": 1.0, | |
| "eval_runtime": 0.2244, | |
| "eval_samples_per_second": 13.369, | |
| "eval_steps_per_second": 13.369, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 11.142857142857142, | |
| "grad_norm": 0.01708046905696392, | |
| "learning_rate": 1.2222222222222222e-05, | |
| "loss": 0.001, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 11.285714285714286, | |
| "grad_norm": 0.025557199493050575, | |
| "learning_rate": 1.1666666666666668e-05, | |
| "loss": 0.0011, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "grad_norm": 0.015760347247123718, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.0008, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 11.571428571428571, | |
| "grad_norm": 0.4625304341316223, | |
| "learning_rate": 1.0555555555555555e-05, | |
| "loss": 0.002, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 11.714285714285714, | |
| "grad_norm": 0.02315388433635235, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0009, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 11.857142857142858, | |
| "grad_norm": 0.013032798655331135, | |
| "learning_rate": 9.444444444444445e-06, | |
| "loss": 0.0006, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.01681060716509819, | |
| "learning_rate": 8.88888888888889e-06, | |
| "loss": 0.0007, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9982174688057041, | |
| "eval_f1": 0.9411764705882353, | |
| "eval_loss": 0.006076255813241005, | |
| "eval_precision": 0.9230769230769231, | |
| "eval_recall": 0.96, | |
| "eval_runtime": 0.2313, | |
| "eval_samples_per_second": 12.968, | |
| "eval_steps_per_second": 12.968, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 12.142857142857142, | |
| "grad_norm": 0.021738462150096893, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.0012, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 12.285714285714286, | |
| "grad_norm": 0.07413913309574127, | |
| "learning_rate": 7.777777777777777e-06, | |
| "loss": 0.001, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 12.428571428571429, | |
| "grad_norm": 0.015694081783294678, | |
| "learning_rate": 7.222222222222222e-06, | |
| "loss": 0.0007, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 12.571428571428571, | |
| "grad_norm": 0.07455814629793167, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.0011, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 12.714285714285714, | |
| "grad_norm": 2.512777805328369, | |
| "learning_rate": 6.111111111111111e-06, | |
| "loss": 0.002, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 12.857142857142858, | |
| "grad_norm": 0.020945778116583824, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0008, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.014726142399013042, | |
| "learning_rate": 5e-06, | |
| "loss": 0.0006, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.9982174688057041, | |
| "eval_f1": 0.9411764705882353, | |
| "eval_loss": 0.010821976698935032, | |
| "eval_precision": 0.9230769230769231, | |
| "eval_recall": 0.96, | |
| "eval_runtime": 0.2328, | |
| "eval_samples_per_second": 12.888, | |
| "eval_steps_per_second": 12.888, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 13.142857142857142, | |
| "grad_norm": 0.022652102634310722, | |
| "learning_rate": 4.444444444444445e-06, | |
| "loss": 0.0007, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 13.285714285714286, | |
| "grad_norm": 0.0192624032497406, | |
| "learning_rate": 3.888888888888889e-06, | |
| "loss": 0.0009, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 13.428571428571429, | |
| "grad_norm": 0.03646668419241905, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.0009, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 13.571428571428571, | |
| "grad_norm": 0.02297368273139, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.0006, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 13.714285714285714, | |
| "grad_norm": 0.013909080997109413, | |
| "learning_rate": 2.2222222222222225e-06, | |
| "loss": 0.0006, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 13.857142857142858, | |
| "grad_norm": 0.021894708275794983, | |
| "learning_rate": 1.6666666666666667e-06, | |
| "loss": 0.0008, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.019053662195801735, | |
| "learning_rate": 1.1111111111111112e-06, | |
| "loss": 0.0011, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9982174688057041, | |
| "eval_f1": 0.9411764705882353, | |
| "eval_loss": 0.011927765794098377, | |
| "eval_precision": 0.9230769230769231, | |
| "eval_recall": 0.96, | |
| "eval_runtime": 0.2367, | |
| "eval_samples_per_second": 12.675, | |
| "eval_steps_per_second": 12.675, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 14.142857142857142, | |
| "grad_norm": 0.015299996361136436, | |
| "learning_rate": 5.555555555555556e-07, | |
| "loss": 0.0007, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 14.285714285714286, | |
| "grad_norm": 0.06155708432197571, | |
| "learning_rate": 0.0, | |
| "loss": 0.0009, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 14.285714285714286, | |
| "eval_accuracy": 0.9982174688057041, | |
| "eval_f1": 0.9411764705882353, | |
| "eval_loss": 0.011938165873289108, | |
| "eval_precision": 0.9230769230769231, | |
| "eval_recall": 0.96, | |
| "eval_runtime": 0.459, | |
| "eval_samples_per_second": 6.535, | |
| "eval_steps_per_second": 6.535, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 52719651225600.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |