| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.111358574610245, | |
| "eval_steps": 25, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0014847809948032665, | |
| "grad_norm": 1.4331752061843872, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.2034, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0014847809948032665, | |
| "eval_loss": 2.588484525680542, | |
| "eval_runtime": 141.9969, | |
| "eval_samples_per_second": 4.0, | |
| "eval_steps_per_second": 2.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.002969561989606533, | |
| "grad_norm": 4.0620622634887695, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.3831, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.004454342984409799, | |
| "grad_norm": 2.3476243019104004, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4242, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.005939123979213066, | |
| "grad_norm": 2.4534640312194824, | |
| "learning_rate": 9.99524110790929e-05, | |
| "loss": 0.2907, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.007423904974016332, | |
| "grad_norm": 1.7481118440628052, | |
| "learning_rate": 9.980973490458728e-05, | |
| "loss": 0.1797, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.008908685968819599, | |
| "grad_norm": 1.0210967063903809, | |
| "learning_rate": 9.957224306869053e-05, | |
| "loss": 0.1305, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.010393466963622866, | |
| "grad_norm": 0.7026821970939636, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.0784, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.011878247958426132, | |
| "grad_norm": 0.8235223293304443, | |
| "learning_rate": 9.881480035599667e-05, | |
| "loss": 0.1005, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.013363028953229399, | |
| "grad_norm": 0.6765594482421875, | |
| "learning_rate": 9.829629131445342e-05, | |
| "loss": 0.0628, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.014847809948032665, | |
| "grad_norm": 1.5128905773162842, | |
| "learning_rate": 9.768584753741134e-05, | |
| "loss": 0.0417, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.016332590942835932, | |
| "grad_norm": 2.1222732067108154, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.1114, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.017817371937639197, | |
| "grad_norm": 0.4752410650253296, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.0758, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.019302152932442463, | |
| "grad_norm": 1.4983466863632202, | |
| "learning_rate": 9.53153893518325e-05, | |
| "loss": 0.0888, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.020786933927245732, | |
| "grad_norm": 0.5699716806411743, | |
| "learning_rate": 9.435054165891109e-05, | |
| "loss": 0.0643, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.022271714922048998, | |
| "grad_norm": 0.8198930025100708, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.0654, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.023756495916852263, | |
| "grad_norm": 0.9171881675720215, | |
| "learning_rate": 9.21695722906443e-05, | |
| "loss": 0.054, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.025241276911655532, | |
| "grad_norm": 0.6817424297332764, | |
| "learning_rate": 9.09576022144496e-05, | |
| "loss": 0.0594, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.026726057906458798, | |
| "grad_norm": 0.9507274627685547, | |
| "learning_rate": 8.966766701456177e-05, | |
| "loss": 0.0351, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.028210838901262063, | |
| "grad_norm": 0.40875244140625, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.0301, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.02969561989606533, | |
| "grad_norm": 0.6699226498603821, | |
| "learning_rate": 8.68638668405062e-05, | |
| "loss": 0.076, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.031180400890868598, | |
| "grad_norm": 0.41454505920410156, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.0228, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.032665181885671864, | |
| "grad_norm": 0.40090444684028625, | |
| "learning_rate": 8.377951038078302e-05, | |
| "loss": 0.0329, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.03414996288047513, | |
| "grad_norm": 0.3448033034801483, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 0.0547, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.035634743875278395, | |
| "grad_norm": 0.5039708614349365, | |
| "learning_rate": 8.043807145043604e-05, | |
| "loss": 0.0548, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.03711952487008166, | |
| "grad_norm": 0.28690701723098755, | |
| "learning_rate": 7.86788218175523e-05, | |
| "loss": 0.0293, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03711952487008166, | |
| "eval_loss": 0.10564589500427246, | |
| "eval_runtime": 143.498, | |
| "eval_samples_per_second": 3.958, | |
| "eval_steps_per_second": 1.979, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.038604305864884926, | |
| "grad_norm": 0.3717527985572815, | |
| "learning_rate": 7.68649804173412e-05, | |
| "loss": 0.0425, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0400890868596882, | |
| "grad_norm": 0.81565260887146, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.0945, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.041573867854491464, | |
| "grad_norm": 0.2406810075044632, | |
| "learning_rate": 7.308743066175172e-05, | |
| "loss": 0.0164, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.04305864884929473, | |
| "grad_norm": 1.2528947591781616, | |
| "learning_rate": 7.113091308703498e-05, | |
| "loss": 0.0452, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.044543429844097995, | |
| "grad_norm": 0.3788902163505554, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.031, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04602821083890126, | |
| "grad_norm": 0.4253796339035034, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.0581, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.047512991833704527, | |
| "grad_norm": 0.251658171415329, | |
| "learning_rate": 6.503528997521366e-05, | |
| "loss": 0.012, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.04899777282850779, | |
| "grad_norm": 0.9965800046920776, | |
| "learning_rate": 6.294095225512603e-05, | |
| "loss": 0.054, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.050482553823311065, | |
| "grad_norm": 0.495941698551178, | |
| "learning_rate": 6.0821980696905146e-05, | |
| "loss": 0.0201, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.05196733481811433, | |
| "grad_norm": 0.4241041839122772, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.0288, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.053452115812917596, | |
| "grad_norm": 0.42201074957847595, | |
| "learning_rate": 5.6526309611002594e-05, | |
| "loss": 0.0273, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.05493689680772086, | |
| "grad_norm": 0.8984993100166321, | |
| "learning_rate": 5.435778713738292e-05, | |
| "loss": 0.0647, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.05642167780252413, | |
| "grad_norm": 0.6671286821365356, | |
| "learning_rate": 5.218096936826681e-05, | |
| "loss": 0.0145, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.05790645879732739, | |
| "grad_norm": 0.07203401625156403, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0018, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.05939123979213066, | |
| "grad_norm": 0.045727696269750595, | |
| "learning_rate": 4.781903063173321e-05, | |
| "loss": 0.0012, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.06087602078693393, | |
| "grad_norm": 0.19696299731731415, | |
| "learning_rate": 4.564221286261709e-05, | |
| "loss": 0.0032, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.062360801781737196, | |
| "grad_norm": 0.3413621485233307, | |
| "learning_rate": 4.347369038899744e-05, | |
| "loss": 0.0059, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.06384558277654045, | |
| "grad_norm": 0.09776332229375839, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.0018, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.06533036377134373, | |
| "grad_norm": 0.0692993700504303, | |
| "learning_rate": 3.917801930309486e-05, | |
| "loss": 0.001, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.066815144766147, | |
| "grad_norm": 0.013437100686132908, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 0.0004, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.06829992576095026, | |
| "grad_norm": 0.9036085605621338, | |
| "learning_rate": 3.4964710024786354e-05, | |
| "loss": 0.0024, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.06978470675575353, | |
| "grad_norm": 14.660072326660156, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 0.4013, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.07126948775055679, | |
| "grad_norm": 2.000504493713379, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.0157, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.07275426874536006, | |
| "grad_norm": 0.019345076754689217, | |
| "learning_rate": 2.886908691296504e-05, | |
| "loss": 0.0004, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.07423904974016332, | |
| "grad_norm": 3.370844841003418, | |
| "learning_rate": 2.6912569338248315e-05, | |
| "loss": 0.2163, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07423904974016332, | |
| "eval_loss": 0.11560910195112228, | |
| "eval_runtime": 143.5027, | |
| "eval_samples_per_second": 3.958, | |
| "eval_steps_per_second": 1.979, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0757238307349666, | |
| "grad_norm": 2.123660087585449, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.1803, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.07720861172976985, | |
| "grad_norm": 2.3457491397857666, | |
| "learning_rate": 2.3135019582658802e-05, | |
| "loss": 0.1052, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.07869339272457312, | |
| "grad_norm": 0.9634860157966614, | |
| "learning_rate": 2.132117818244771e-05, | |
| "loss": 0.1048, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.0801781737193764, | |
| "grad_norm": 1.2893617153167725, | |
| "learning_rate": 1.9561928549563968e-05, | |
| "loss": 0.1053, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.08166295471417966, | |
| "grad_norm": 1.0800879001617432, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.1001, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.08314773570898293, | |
| "grad_norm": 1.4931827783584595, | |
| "learning_rate": 1.622048961921699e-05, | |
| "loss": 0.0806, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.08463251670378619, | |
| "grad_norm": 3.529783248901367, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.1045, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.08611729769858946, | |
| "grad_norm": 2.6875853538513184, | |
| "learning_rate": 1.3136133159493802e-05, | |
| "loss": 0.0877, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.08760207869339272, | |
| "grad_norm": 1.4070323705673218, | |
| "learning_rate": 1.1697777844051105e-05, | |
| "loss": 0.0469, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.08908685968819599, | |
| "grad_norm": 0.904192328453064, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 0.037, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.09057164068299926, | |
| "grad_norm": 1.2252767086029053, | |
| "learning_rate": 9.042397785550405e-06, | |
| "loss": 0.0616, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.09205642167780252, | |
| "grad_norm": 0.8113102912902832, | |
| "learning_rate": 7.830427709355725e-06, | |
| "loss": 0.0507, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.0935412026726058, | |
| "grad_norm": 0.6455811262130737, | |
| "learning_rate": 6.698729810778065e-06, | |
| "loss": 0.0617, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.09502598366740905, | |
| "grad_norm": 0.48664721846580505, | |
| "learning_rate": 5.649458341088915e-06, | |
| "loss": 0.0352, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.09651076466221233, | |
| "grad_norm": 3.366154909133911, | |
| "learning_rate": 4.684610648167503e-06, | |
| "loss": 0.0314, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.09799554565701558, | |
| "grad_norm": 1.3770588636398315, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0382, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.09948032665181886, | |
| "grad_norm": 0.6862576007843018, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.0242, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.10096510764662213, | |
| "grad_norm": 0.8646849393844604, | |
| "learning_rate": 2.314152462588659e-06, | |
| "loss": 0.0473, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.10244988864142539, | |
| "grad_norm": 0.4752139151096344, | |
| "learning_rate": 1.70370868554659e-06, | |
| "loss": 0.0625, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.10393466963622866, | |
| "grad_norm": 1.9038113355636597, | |
| "learning_rate": 1.1851996440033319e-06, | |
| "loss": 0.051, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.10541945063103192, | |
| "grad_norm": 0.5322926044464111, | |
| "learning_rate": 7.596123493895991e-07, | |
| "loss": 0.0765, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.10690423162583519, | |
| "grad_norm": 0.49304574728012085, | |
| "learning_rate": 4.277569313094809e-07, | |
| "loss": 0.0655, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.10838901262063845, | |
| "grad_norm": 0.17058974504470825, | |
| "learning_rate": 1.9026509541272275e-07, | |
| "loss": 0.027, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.10987379361544172, | |
| "grad_norm": 0.644688606262207, | |
| "learning_rate": 4.7588920907110094e-08, | |
| "loss": 0.0848, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.111358574610245, | |
| "grad_norm": 0.3389902114868164, | |
| "learning_rate": 0.0, | |
| "loss": 0.0288, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.111358574610245, | |
| "eval_loss": 0.11129310727119446, | |
| "eval_runtime": 143.6519, | |
| "eval_samples_per_second": 3.954, | |
| "eval_steps_per_second": 1.977, | |
| "step": 75 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 75, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.1208807043773235e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |