| { | |
| "best_metric": 1.2658852338790894, | |
| "best_model_checkpoint": "data/Llama-31-8B_task-1_120-samples_config-3/checkpoint-198", | |
| "epoch": 25.0, | |
| "eval_steps": 500, | |
| "global_step": 275, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09090909090909091, | |
| "grad_norm": 1.9471415281295776, | |
| "learning_rate": 6.060606060606061e-08, | |
| "loss": 2.2569, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.18181818181818182, | |
| "grad_norm": 1.9496530294418335, | |
| "learning_rate": 1.2121212121212122e-07, | |
| "loss": 2.3228, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 1.717644453048706, | |
| "learning_rate": 2.4242424242424244e-07, | |
| "loss": 2.0831, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.5454545454545454, | |
| "grad_norm": 1.5924347639083862, | |
| "learning_rate": 3.6363636363636366e-07, | |
| "loss": 2.0746, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 1.8148092031478882, | |
| "learning_rate": 4.848484848484849e-07, | |
| "loss": 2.1661, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "grad_norm": 1.6654292345046997, | |
| "learning_rate": 6.060606060606061e-07, | |
| "loss": 1.9625, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.0993740558624268, | |
| "eval_runtime": 9.6328, | |
| "eval_samples_per_second": 2.491, | |
| "eval_steps_per_second": 2.491, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 1.543703556060791, | |
| "learning_rate": 7.272727272727273e-07, | |
| "loss": 1.9298, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.2727272727272727, | |
| "grad_norm": 1.8203456401824951, | |
| "learning_rate": 8.484848484848486e-07, | |
| "loss": 2.1289, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "grad_norm": 1.7504804134368896, | |
| "learning_rate": 9.696969696969698e-07, | |
| "loss": 2.153, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6363636363636362, | |
| "grad_norm": 1.895839810371399, | |
| "learning_rate": 1.090909090909091e-06, | |
| "loss": 2.0811, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 1.7807648181915283, | |
| "learning_rate": 1.2121212121212122e-06, | |
| "loss": 2.0446, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.8615657091140747, | |
| "learning_rate": 1.3333333333333334e-06, | |
| "loss": 2.1365, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.0816335678100586, | |
| "eval_runtime": 9.6405, | |
| "eval_samples_per_second": 2.49, | |
| "eval_steps_per_second": 2.49, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "grad_norm": 1.8219581842422485, | |
| "learning_rate": 1.4545454545454546e-06, | |
| "loss": 2.1017, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.3636363636363638, | |
| "grad_norm": 1.6915614604949951, | |
| "learning_rate": 1.5757575757575759e-06, | |
| "loss": 2.0744, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "grad_norm": 1.8501787185668945, | |
| "learning_rate": 1.6969696969696973e-06, | |
| "loss": 2.0474, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 1.6685246229171753, | |
| "learning_rate": 1.8181818181818183e-06, | |
| "loss": 1.9084, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "grad_norm": 2.073890209197998, | |
| "learning_rate": 1.9393939393939395e-06, | |
| "loss": 2.1371, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.0466740131378174, | |
| "eval_runtime": 9.6342, | |
| "eval_samples_per_second": 2.491, | |
| "eval_steps_per_second": 2.491, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.090909090909091, | |
| "grad_norm": 1.7652820348739624, | |
| "learning_rate": 2.0606060606060607e-06, | |
| "loss": 2.1285, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.2727272727272725, | |
| "grad_norm": 2.0258305072784424, | |
| "learning_rate": 2.181818181818182e-06, | |
| "loss": 2.0643, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.4545454545454546, | |
| "grad_norm": 1.8539706468582153, | |
| "learning_rate": 2.303030303030303e-06, | |
| "loss": 1.9892, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 1.8331531286239624, | |
| "learning_rate": 2.4242424242424244e-06, | |
| "loss": 1.9928, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.8181818181818183, | |
| "grad_norm": 1.7377302646636963, | |
| "learning_rate": 2.5454545454545456e-06, | |
| "loss": 1.9357, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.0487308502197266, | |
| "learning_rate": 2.666666666666667e-06, | |
| "loss": 2.0536, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.9862397909164429, | |
| "eval_runtime": 9.6288, | |
| "eval_samples_per_second": 2.493, | |
| "eval_steps_per_second": 2.493, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.181818181818182, | |
| "grad_norm": 1.935196876525879, | |
| "learning_rate": 2.7878787878787885e-06, | |
| "loss": 1.9591, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.363636363636363, | |
| "grad_norm": 1.7882719039916992, | |
| "learning_rate": 2.9090909090909093e-06, | |
| "loss": 1.9446, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.545454545454545, | |
| "grad_norm": 1.80415940284729, | |
| "learning_rate": 3.0303030303030305e-06, | |
| "loss": 1.9487, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 1.7832120656967163, | |
| "learning_rate": 3.1515151515151517e-06, | |
| "loss": 1.9723, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 4.909090909090909, | |
| "grad_norm": 1.6389567852020264, | |
| "learning_rate": 3.272727272727273e-06, | |
| "loss": 1.8317, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.8956037759780884, | |
| "eval_runtime": 9.6308, | |
| "eval_samples_per_second": 2.492, | |
| "eval_steps_per_second": 2.492, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.090909090909091, | |
| "grad_norm": 1.7195862531661987, | |
| "learning_rate": 3.3939393939393946e-06, | |
| "loss": 1.9008, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.2727272727272725, | |
| "grad_norm": 1.8558402061462402, | |
| "learning_rate": 3.5151515151515154e-06, | |
| "loss": 1.916, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.454545454545454, | |
| "grad_norm": 1.9300419092178345, | |
| "learning_rate": 3.6363636363636366e-06, | |
| "loss": 1.8262, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.636363636363637, | |
| "grad_norm": 1.8043519258499146, | |
| "learning_rate": 3.757575757575758e-06, | |
| "loss": 1.8013, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 5.818181818181818, | |
| "grad_norm": 1.5544039011001587, | |
| "learning_rate": 3.878787878787879e-06, | |
| "loss": 1.7996, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.6001099348068237, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.7607, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.7668356895446777, | |
| "eval_runtime": 9.6388, | |
| "eval_samples_per_second": 2.49, | |
| "eval_steps_per_second": 2.49, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.181818181818182, | |
| "grad_norm": 1.52090322971344, | |
| "learning_rate": 4.1212121212121215e-06, | |
| "loss": 1.7673, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.363636363636363, | |
| "grad_norm": 1.4516123533248901, | |
| "learning_rate": 4.242424242424243e-06, | |
| "loss": 1.782, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.545454545454545, | |
| "grad_norm": 1.318569540977478, | |
| "learning_rate": 4.363636363636364e-06, | |
| "loss": 1.6176, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 6.7272727272727275, | |
| "grad_norm": 1.3474187850952148, | |
| "learning_rate": 4.4848484848484855e-06, | |
| "loss": 1.6543, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "grad_norm": 1.1612062454223633, | |
| "learning_rate": 4.606060606060606e-06, | |
| "loss": 1.6452, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 1.6453262567520142, | |
| "eval_runtime": 9.6333, | |
| "eval_samples_per_second": 2.491, | |
| "eval_steps_per_second": 2.491, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 7.090909090909091, | |
| "grad_norm": 1.062233328819275, | |
| "learning_rate": 4.727272727272728e-06, | |
| "loss": 1.5258, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 7.2727272727272725, | |
| "grad_norm": 0.9276080131530762, | |
| "learning_rate": 4.848484848484849e-06, | |
| "loss": 1.7018, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 7.454545454545454, | |
| "grad_norm": 0.7713845372200012, | |
| "learning_rate": 4.9696969696969696e-06, | |
| "loss": 1.5366, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 7.636363636363637, | |
| "grad_norm": 0.8294752240180969, | |
| "learning_rate": 5.090909090909091e-06, | |
| "loss": 1.5103, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 7.818181818181818, | |
| "grad_norm": 0.7149582505226135, | |
| "learning_rate": 5.212121212121213e-06, | |
| "loss": 1.5768, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.7764841914176941, | |
| "learning_rate": 5.333333333333334e-06, | |
| "loss": 1.548, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.5728096961975098, | |
| "eval_runtime": 9.6306, | |
| "eval_samples_per_second": 2.492, | |
| "eval_steps_per_second": 2.492, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 8.181818181818182, | |
| "grad_norm": 0.7410159111022949, | |
| "learning_rate": 5.4545454545454545e-06, | |
| "loss": 1.4837, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 8.363636363636363, | |
| "grad_norm": 0.7778185606002808, | |
| "learning_rate": 5.575757575757577e-06, | |
| "loss": 1.518, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 8.545454545454545, | |
| "grad_norm": 0.7047359943389893, | |
| "learning_rate": 5.696969696969698e-06, | |
| "loss": 1.5698, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 8.727272727272727, | |
| "grad_norm": 0.8407279253005981, | |
| "learning_rate": 5.8181818181818185e-06, | |
| "loss": 1.5277, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 8.909090909090908, | |
| "grad_norm": 0.6757096648216248, | |
| "learning_rate": 5.93939393939394e-06, | |
| "loss": 1.4631, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 1.5217022895812988, | |
| "eval_runtime": 9.631, | |
| "eval_samples_per_second": 2.492, | |
| "eval_steps_per_second": 2.492, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 9.090909090909092, | |
| "grad_norm": 0.6718485355377197, | |
| "learning_rate": 6.060606060606061e-06, | |
| "loss": 1.4975, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 9.272727272727273, | |
| "grad_norm": 0.657922089099884, | |
| "learning_rate": 6.181818181818182e-06, | |
| "loss": 1.457, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 9.454545454545455, | |
| "grad_norm": 0.8073777556419373, | |
| "learning_rate": 6.303030303030303e-06, | |
| "loss": 1.5072, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 9.636363636363637, | |
| "grad_norm": 0.7699378132820129, | |
| "learning_rate": 6.424242424242425e-06, | |
| "loss": 1.4021, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 9.818181818181818, | |
| "grad_norm": 0.7344899773597717, | |
| "learning_rate": 6.545454545454546e-06, | |
| "loss": 1.4539, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.6962461471557617, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 1.4126, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.471091866493225, | |
| "eval_runtime": 9.628, | |
| "eval_samples_per_second": 2.493, | |
| "eval_steps_per_second": 2.493, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 10.181818181818182, | |
| "grad_norm": 0.756142258644104, | |
| "learning_rate": 6.787878787878789e-06, | |
| "loss": 1.4482, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 10.363636363636363, | |
| "grad_norm": 0.7393659353256226, | |
| "learning_rate": 6.90909090909091e-06, | |
| "loss": 1.45, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 10.545454545454545, | |
| "grad_norm": 0.7522909641265869, | |
| "learning_rate": 7.030303030303031e-06, | |
| "loss": 1.4183, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 10.727272727272727, | |
| "grad_norm": 0.632103681564331, | |
| "learning_rate": 7.151515151515152e-06, | |
| "loss": 1.4051, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "grad_norm": 0.6698535680770874, | |
| "learning_rate": 7.272727272727273e-06, | |
| "loss": 1.3079, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 1.4176009893417358, | |
| "eval_runtime": 9.6277, | |
| "eval_samples_per_second": 2.493, | |
| "eval_steps_per_second": 2.493, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 11.090909090909092, | |
| "grad_norm": 0.7236391305923462, | |
| "learning_rate": 7.393939393939395e-06, | |
| "loss": 1.3409, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 11.272727272727273, | |
| "grad_norm": 0.7769017219543457, | |
| "learning_rate": 7.515151515151516e-06, | |
| "loss": 1.3852, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 11.454545454545455, | |
| "grad_norm": 0.7625659704208374, | |
| "learning_rate": 7.636363636363638e-06, | |
| "loss": 1.4075, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 11.636363636363637, | |
| "grad_norm": 0.7391623258590698, | |
| "learning_rate": 7.757575757575758e-06, | |
| "loss": 1.3438, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 11.818181818181818, | |
| "grad_norm": 0.7174716591835022, | |
| "learning_rate": 7.87878787878788e-06, | |
| "loss": 1.2229, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.6933201551437378, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.3012, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 1.3768787384033203, | |
| "eval_runtime": 9.6401, | |
| "eval_samples_per_second": 2.49, | |
| "eval_steps_per_second": 2.49, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 12.181818181818182, | |
| "grad_norm": 0.7602589130401611, | |
| "learning_rate": 8.121212121212121e-06, | |
| "loss": 1.3075, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 12.363636363636363, | |
| "grad_norm": 0.7896175384521484, | |
| "learning_rate": 8.242424242424243e-06, | |
| "loss": 1.2531, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 12.545454545454545, | |
| "grad_norm": 0.7628654837608337, | |
| "learning_rate": 8.363636363636365e-06, | |
| "loss": 1.2487, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 12.727272727272727, | |
| "grad_norm": 0.7553589940071106, | |
| "learning_rate": 8.484848484848486e-06, | |
| "loss": 1.3375, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 12.909090909090908, | |
| "grad_norm": 0.8193318247795105, | |
| "learning_rate": 8.606060606060606e-06, | |
| "loss": 1.2575, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 1.34230375289917, | |
| "eval_runtime": 9.6362, | |
| "eval_samples_per_second": 2.491, | |
| "eval_steps_per_second": 2.491, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 13.090909090909092, | |
| "grad_norm": 0.7942626476287842, | |
| "learning_rate": 8.727272727272728e-06, | |
| "loss": 1.2746, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 13.272727272727273, | |
| "grad_norm": 0.7456735372543335, | |
| "learning_rate": 8.84848484848485e-06, | |
| "loss": 1.2148, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 13.454545454545455, | |
| "grad_norm": 0.8296701908111572, | |
| "learning_rate": 8.969696969696971e-06, | |
| "loss": 1.2372, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 13.636363636363637, | |
| "grad_norm": 0.8604351282119751, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 1.1917, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 13.818181818181818, | |
| "grad_norm": 0.8084425330162048, | |
| "learning_rate": 9.212121212121213e-06, | |
| "loss": 1.2257, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.8843783140182495, | |
| "learning_rate": 9.333333333333334e-06, | |
| "loss": 1.2537, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 1.3097882270812988, | |
| "eval_runtime": 9.642, | |
| "eval_samples_per_second": 2.489, | |
| "eval_steps_per_second": 2.489, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 14.181818181818182, | |
| "grad_norm": 0.9855999946594238, | |
| "learning_rate": 9.454545454545456e-06, | |
| "loss": 1.1807, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 14.363636363636363, | |
| "grad_norm": 0.7823004722595215, | |
| "learning_rate": 9.575757575757576e-06, | |
| "loss": 1.1481, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 14.545454545454545, | |
| "grad_norm": 0.8486412167549133, | |
| "learning_rate": 9.696969696969698e-06, | |
| "loss": 1.1524, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 14.727272727272727, | |
| "grad_norm": 0.9296090006828308, | |
| "learning_rate": 9.81818181818182e-06, | |
| "loss": 1.223, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 14.909090909090908, | |
| "grad_norm": 0.962130069732666, | |
| "learning_rate": 9.939393939393939e-06, | |
| "loss": 1.1994, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 1.2874382734298706, | |
| "eval_runtime": 9.6369, | |
| "eval_samples_per_second": 2.49, | |
| "eval_steps_per_second": 2.49, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 15.090909090909092, | |
| "grad_norm": 1.0040925741195679, | |
| "learning_rate": 9.999988811118232e-06, | |
| "loss": 1.0752, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 15.272727272727273, | |
| "grad_norm": 0.8788989782333374, | |
| "learning_rate": 9.999899300364534e-06, | |
| "loss": 1.1945, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 15.454545454545455, | |
| "grad_norm": 0.8372966647148132, | |
| "learning_rate": 9.999720280459576e-06, | |
| "loss": 1.1612, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 15.636363636363637, | |
| "grad_norm": 0.9981568455696106, | |
| "learning_rate": 9.999451754608208e-06, | |
| "loss": 1.0675, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 15.818181818181818, | |
| "grad_norm": 0.989941418170929, | |
| "learning_rate": 9.99909372761763e-06, | |
| "loss": 1.1123, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 1.071035623550415, | |
| "learning_rate": 9.99864620589731e-06, | |
| "loss": 1.1054, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 1.2712817192077637, | |
| "eval_runtime": 9.6277, | |
| "eval_samples_per_second": 2.493, | |
| "eval_steps_per_second": 2.493, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 16.181818181818183, | |
| "grad_norm": 1.0144329071044922, | |
| "learning_rate": 9.998109197458865e-06, | |
| "loss": 1.1681, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 16.363636363636363, | |
| "grad_norm": 1.0573251247406006, | |
| "learning_rate": 9.997482711915926e-06, | |
| "loss": 1.06, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 16.545454545454547, | |
| "grad_norm": 1.000741720199585, | |
| "learning_rate": 9.996766760483955e-06, | |
| "loss": 1.0508, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 16.727272727272727, | |
| "grad_norm": 1.2005733251571655, | |
| "learning_rate": 9.995961355980052e-06, | |
| "loss": 1.0129, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 16.90909090909091, | |
| "grad_norm": 1.195275902748108, | |
| "learning_rate": 9.99506651282272e-06, | |
| "loss": 1.0452, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 1.2680319547653198, | |
| "eval_runtime": 9.6377, | |
| "eval_samples_per_second": 2.49, | |
| "eval_steps_per_second": 2.49, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 17.09090909090909, | |
| "grad_norm": 1.2155280113220215, | |
| "learning_rate": 9.994082247031613e-06, | |
| "loss": 1.0263, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 17.272727272727273, | |
| "grad_norm": 1.143007516860962, | |
| "learning_rate": 9.993008576227248e-06, | |
| "loss": 0.9197, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 17.454545454545453, | |
| "grad_norm": 1.260664939880371, | |
| "learning_rate": 9.991845519630679e-06, | |
| "loss": 0.9681, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 17.636363636363637, | |
| "grad_norm": 1.3235067129135132, | |
| "learning_rate": 9.99059309806317e-06, | |
| "loss": 1.0343, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 17.818181818181817, | |
| "grad_norm": 1.4046649932861328, | |
| "learning_rate": 9.989251333945813e-06, | |
| "loss": 1.0078, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 1.2338204383850098, | |
| "learning_rate": 9.987820251299121e-06, | |
| "loss": 1.0716, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 1.2658852338790894, | |
| "eval_runtime": 9.6422, | |
| "eval_samples_per_second": 2.489, | |
| "eval_steps_per_second": 2.489, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 18.181818181818183, | |
| "grad_norm": 1.3795467615127563, | |
| "learning_rate": 9.986299875742612e-06, | |
| "loss": 0.9914, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 18.363636363636363, | |
| "grad_norm": 1.2211111783981323, | |
| "learning_rate": 9.984690234494338e-06, | |
| "loss": 0.9788, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 18.545454545454547, | |
| "grad_norm": 1.4873826503753662, | |
| "learning_rate": 9.982991356370404e-06, | |
| "loss": 0.9206, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 18.727272727272727, | |
| "grad_norm": 1.1939488649368286, | |
| "learning_rate": 9.98120327178445e-06, | |
| "loss": 0.9713, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 18.90909090909091, | |
| "grad_norm": 1.4163975715637207, | |
| "learning_rate": 9.979326012747106e-06, | |
| "loss": 0.9207, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 1.2754586935043335, | |
| "eval_runtime": 9.6506, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 19.09090909090909, | |
| "grad_norm": 1.57706880569458, | |
| "learning_rate": 9.977359612865424e-06, | |
| "loss": 0.847, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 19.272727272727273, | |
| "grad_norm": 1.3526296615600586, | |
| "learning_rate": 9.975304107342268e-06, | |
| "loss": 0.83, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 19.454545454545453, | |
| "grad_norm": 1.3322209119796753, | |
| "learning_rate": 9.973159532975691e-06, | |
| "loss": 0.9059, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 19.636363636363637, | |
| "grad_norm": 1.7349506616592407, | |
| "learning_rate": 9.970925928158275e-06, | |
| "loss": 0.9101, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 19.818181818181817, | |
| "grad_norm": 1.7136040925979614, | |
| "learning_rate": 9.968603332876435e-06, | |
| "loss": 0.9291, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 1.6715441942214966, | |
| "learning_rate": 9.966191788709716e-06, | |
| "loss": 0.8712, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 1.291751503944397, | |
| "eval_runtime": 9.6384, | |
| "eval_samples_per_second": 2.49, | |
| "eval_steps_per_second": 2.49, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 20.181818181818183, | |
| "grad_norm": 1.7461074590682983, | |
| "learning_rate": 9.963691338830045e-06, | |
| "loss": 0.8728, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 20.363636363636363, | |
| "grad_norm": 1.5934933423995972, | |
| "learning_rate": 9.961102028000948e-06, | |
| "loss": 0.7777, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 20.545454545454547, | |
| "grad_norm": 1.8316841125488281, | |
| "learning_rate": 9.958423902576764e-06, | |
| "loss": 0.9123, | |
| "step": 226 | |
| }, | |
| { | |
| "epoch": 20.727272727272727, | |
| "grad_norm": 1.6827653646469116, | |
| "learning_rate": 9.955657010501807e-06, | |
| "loss": 0.741, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 20.90909090909091, | |
| "grad_norm": 1.7905974388122559, | |
| "learning_rate": 9.952801401309504e-06, | |
| "loss": 0.8179, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_loss": 1.337079644203186, | |
| "eval_runtime": 9.6484, | |
| "eval_samples_per_second": 2.487, | |
| "eval_steps_per_second": 2.487, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 21.09090909090909, | |
| "grad_norm": 1.711150884628296, | |
| "learning_rate": 9.949857126121519e-06, | |
| "loss": 0.7859, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 21.272727272727273, | |
| "grad_norm": 1.8518304824829102, | |
| "learning_rate": 9.946824237646823e-06, | |
| "loss": 0.7743, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 21.454545454545453, | |
| "grad_norm": 2.065861701965332, | |
| "learning_rate": 9.94370279018077e-06, | |
| "loss": 0.7725, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 21.636363636363637, | |
| "grad_norm": 1.9285728931427002, | |
| "learning_rate": 9.940492839604103e-06, | |
| "loss": 0.7043, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 21.818181818181817, | |
| "grad_norm": 1.9804288148880005, | |
| "learning_rate": 9.937194443381972e-06, | |
| "loss": 0.8355, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "grad_norm": 1.9706605672836304, | |
| "learning_rate": 9.933807660562898e-06, | |
| "loss": 0.6485, | |
| "step": 242 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 1.3560718297958374, | |
| "eval_runtime": 9.6266, | |
| "eval_samples_per_second": 2.493, | |
| "eval_steps_per_second": 2.493, | |
| "step": 242 | |
| }, | |
| { | |
| "epoch": 22.181818181818183, | |
| "grad_norm": 1.8734219074249268, | |
| "learning_rate": 9.930332551777709e-06, | |
| "loss": 0.6863, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 22.363636363636363, | |
| "grad_norm": 2.110110282897949, | |
| "learning_rate": 9.926769179238467e-06, | |
| "loss": 0.6799, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 22.545454545454547, | |
| "grad_norm": 2.442317008972168, | |
| "learning_rate": 9.923117606737347e-06, | |
| "loss": 0.6703, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 22.727272727272727, | |
| "grad_norm": 2.216526508331299, | |
| "learning_rate": 9.919377899645497e-06, | |
| "loss": 0.687, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 22.90909090909091, | |
| "grad_norm": 2.3219642639160156, | |
| "learning_rate": 9.915550124911866e-06, | |
| "loss": 0.6958, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_loss": 1.4414390325546265, | |
| "eval_runtime": 9.6326, | |
| "eval_samples_per_second": 2.492, | |
| "eval_steps_per_second": 2.492, | |
| "step": 253 | |
| }, | |
| { | |
| "epoch": 23.09090909090909, | |
| "grad_norm": 2.1852304935455322, | |
| "learning_rate": 9.91163435106201e-06, | |
| "loss": 0.5976, | |
| "step": 254 | |
| }, | |
| { | |
| "epoch": 23.272727272727273, | |
| "grad_norm": 2.0445878505706787, | |
| "learning_rate": 9.907630648196857e-06, | |
| "loss": 0.687, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 23.454545454545453, | |
| "grad_norm": 2.2275197505950928, | |
| "learning_rate": 9.903539087991462e-06, | |
| "loss": 0.62, | |
| "step": 258 | |
| }, | |
| { | |
| "epoch": 23.636363636363637, | |
| "grad_norm": 2.462937593460083, | |
| "learning_rate": 9.899359743693715e-06, | |
| "loss": 0.5359, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 23.818181818181817, | |
| "grad_norm": 2.7242631912231445, | |
| "learning_rate": 9.895092690123036e-06, | |
| "loss": 0.6217, | |
| "step": 262 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 2.585885763168335, | |
| "learning_rate": 9.890738003669029e-06, | |
| "loss": 0.5845, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 1.5147289037704468, | |
| "eval_runtime": 9.6323, | |
| "eval_samples_per_second": 2.492, | |
| "eval_steps_per_second": 2.492, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 24.181818181818183, | |
| "grad_norm": 2.6411542892456055, | |
| "learning_rate": 9.886295762290125e-06, | |
| "loss": 0.5457, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 24.363636363636363, | |
| "grad_norm": 2.4892776012420654, | |
| "learning_rate": 9.881766045512176e-06, | |
| "loss": 0.5765, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 24.545454545454547, | |
| "grad_norm": 2.6362385749816895, | |
| "learning_rate": 9.877148934427037e-06, | |
| "loss": 0.5477, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 24.727272727272727, | |
| "grad_norm": 2.6221413612365723, | |
| "learning_rate": 9.872444511691108e-06, | |
| "loss": 0.4876, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 24.90909090909091, | |
| "grad_norm": 3.3728928565979004, | |
| "learning_rate": 9.867652861523866e-06, | |
| "loss": 0.5274, | |
| "step": 274 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 1.5911802053451538, | |
| "eval_runtime": 9.6269, | |
| "eval_samples_per_second": 2.493, | |
| "eval_steps_per_second": 2.493, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "step": 275, | |
| "total_flos": 3.3769591557586944e+16, | |
| "train_loss": 1.3221848821640014, | |
| "train_runtime": 2580.936, | |
| "train_samples_per_second": 5.114, | |
| "train_steps_per_second": 0.639 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 1650, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 150, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 7, | |
| "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": 3.3769591557586944e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |