| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.10621348911311737, | |
| "eval_steps": 25, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0014161798548415648, | |
| "grad_norm": 23.20075035095215, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 11.3912, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0014161798548415648, | |
| "eval_loss": 10.59463119506836, | |
| "eval_runtime": 151.6501, | |
| "eval_samples_per_second": 3.924, | |
| "eval_steps_per_second": 1.965, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0028323597096831295, | |
| "grad_norm": 28.277725219726562, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 10.9081, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.004248539564524694, | |
| "grad_norm": 24.364511489868164, | |
| "learning_rate": 0.0001, | |
| "loss": 10.0504, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.005664719419366259, | |
| "grad_norm": 18.489341735839844, | |
| "learning_rate": 9.99524110790929e-05, | |
| "loss": 7.3342, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.007080899274207825, | |
| "grad_norm": 25.614126205444336, | |
| "learning_rate": 9.980973490458728e-05, | |
| "loss": 3.8967, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.008497079129049389, | |
| "grad_norm": 41.44906234741211, | |
| "learning_rate": 9.957224306869053e-05, | |
| "loss": 1.9891, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.009913258983890954, | |
| "grad_norm": 24.81721305847168, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 1.3137, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.011329438838732518, | |
| "grad_norm": 7.7688493728637695, | |
| "learning_rate": 9.881480035599667e-05, | |
| "loss": 0.4025, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.012745618693574084, | |
| "grad_norm": 4.148148059844971, | |
| "learning_rate": 9.829629131445342e-05, | |
| "loss": 0.1716, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.01416179854841565, | |
| "grad_norm": 0.756351113319397, | |
| "learning_rate": 9.768584753741134e-05, | |
| "loss": 0.0929, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.015577978403257213, | |
| "grad_norm": 4.284754276275635, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.1716, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.016994158258098777, | |
| "grad_norm": 5.088748455047607, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.1915, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.018410338112940343, | |
| "grad_norm": 3.0917534828186035, | |
| "learning_rate": 9.53153893518325e-05, | |
| "loss": 0.1343, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.01982651796778191, | |
| "grad_norm": 3.725177049636841, | |
| "learning_rate": 9.435054165891109e-05, | |
| "loss": 0.1964, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.021242697822623474, | |
| "grad_norm": 1.8048001527786255, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.1206, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.022658877677465036, | |
| "grad_norm": 0.94793301820755, | |
| "learning_rate": 9.21695722906443e-05, | |
| "loss": 0.1201, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.024075057532306602, | |
| "grad_norm": 2.34434175491333, | |
| "learning_rate": 9.09576022144496e-05, | |
| "loss": 0.1184, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.025491237387148168, | |
| "grad_norm": 2.191009759902954, | |
| "learning_rate": 8.966766701456177e-05, | |
| "loss": 0.1479, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.026907417241989733, | |
| "grad_norm": 1.4999676942825317, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.1267, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0283235970968313, | |
| "grad_norm": 0.44820383191108704, | |
| "learning_rate": 8.68638668405062e-05, | |
| "loss": 0.1192, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02973977695167286, | |
| "grad_norm": 2.531404733657837, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.1483, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.031155956806514427, | |
| "grad_norm": 0.7770451903343201, | |
| "learning_rate": 8.377951038078302e-05, | |
| "loss": 0.1114, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.03257213666135599, | |
| "grad_norm": 1.0737981796264648, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 0.118, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.033988316516197555, | |
| "grad_norm": 3.9276962280273438, | |
| "learning_rate": 8.043807145043604e-05, | |
| "loss": 0.2074, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.035404496371039124, | |
| "grad_norm": 4.217297554016113, | |
| "learning_rate": 7.86788218175523e-05, | |
| "loss": 0.2623, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.035404496371039124, | |
| "eval_loss": 0.1676926612854004, | |
| "eval_runtime": 152.4912, | |
| "eval_samples_per_second": 3.902, | |
| "eval_steps_per_second": 1.954, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.036820676225880686, | |
| "grad_norm": 2.2406327724456787, | |
| "learning_rate": 7.68649804173412e-05, | |
| "loss": 0.1696, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.038236856080722255, | |
| "grad_norm": 0.21706989407539368, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.1207, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.03965303593556382, | |
| "grad_norm": 0.6819403767585754, | |
| "learning_rate": 7.308743066175172e-05, | |
| "loss": 0.1059, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.04106921579040538, | |
| "grad_norm": 1.85183584690094, | |
| "learning_rate": 7.113091308703498e-05, | |
| "loss": 0.1389, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.04248539564524695, | |
| "grad_norm": 0.6912545561790466, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.1074, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04390157550008851, | |
| "grad_norm": 0.2985907196998596, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.1139, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.04531775535493007, | |
| "grad_norm": 0.9557470083236694, | |
| "learning_rate": 6.503528997521366e-05, | |
| "loss": 0.1193, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.04673393520977164, | |
| "grad_norm": 0.38003310561180115, | |
| "learning_rate": 6.294095225512603e-05, | |
| "loss": 0.112, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.048150115064613204, | |
| "grad_norm": 0.3061525523662567, | |
| "learning_rate": 6.0821980696905146e-05, | |
| "loss": 0.107, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.04956629491945477, | |
| "grad_norm": 0.29527008533477783, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.1145, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.050982474774296335, | |
| "grad_norm": 0.4223112463951111, | |
| "learning_rate": 5.6526309611002594e-05, | |
| "loss": 0.0935, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0523986546291379, | |
| "grad_norm": 0.8453797101974487, | |
| "learning_rate": 5.435778713738292e-05, | |
| "loss": 0.0726, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.053814834483979467, | |
| "grad_norm": 0.26757749915122986, | |
| "learning_rate": 5.218096936826681e-05, | |
| "loss": 0.0648, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.05523101433882103, | |
| "grad_norm": 1.951507806777954, | |
| "learning_rate": 5e-05, | |
| "loss": 0.145, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0566471941936626, | |
| "grad_norm": 2.7410080432891846, | |
| "learning_rate": 4.781903063173321e-05, | |
| "loss": 0.206, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05806337404850416, | |
| "grad_norm": 0.6865355968475342, | |
| "learning_rate": 4.564221286261709e-05, | |
| "loss": 0.0908, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.05947955390334572, | |
| "grad_norm": 0.949466347694397, | |
| "learning_rate": 4.347369038899744e-05, | |
| "loss": 0.0796, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.06089573375818729, | |
| "grad_norm": 0.7530195713043213, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.0922, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.06231191361302885, | |
| "grad_norm": 1.2768175601959229, | |
| "learning_rate": 3.917801930309486e-05, | |
| "loss": 0.0763, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.06372809346787042, | |
| "grad_norm": 0.44944190979003906, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 0.0563, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.06514427332271198, | |
| "grad_norm": 1.1507816314697266, | |
| "learning_rate": 3.4964710024786354e-05, | |
| "loss": 0.1047, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.06656045317755355, | |
| "grad_norm": 0.27491295337677, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 0.0581, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.06797663303239511, | |
| "grad_norm": 0.8990926742553711, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.1178, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.06939281288723669, | |
| "grad_norm": 0.36759093403816223, | |
| "learning_rate": 2.886908691296504e-05, | |
| "loss": 0.0581, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.07080899274207825, | |
| "grad_norm": 0.25813916325569153, | |
| "learning_rate": 2.6912569338248315e-05, | |
| "loss": 0.0469, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07080899274207825, | |
| "eval_loss": 0.10698376595973969, | |
| "eval_runtime": 153.0381, | |
| "eval_samples_per_second": 3.888, | |
| "eval_steps_per_second": 1.947, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07222517259691981, | |
| "grad_norm": 6.394037246704102, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.752, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.07364135245176137, | |
| "grad_norm": 0.4659101068973541, | |
| "learning_rate": 2.3135019582658802e-05, | |
| "loss": 0.0882, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.07505753230660293, | |
| "grad_norm": 0.36920779943466187, | |
| "learning_rate": 2.132117818244771e-05, | |
| "loss": 0.0927, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.07647371216144451, | |
| "grad_norm": 0.7038587927818298, | |
| "learning_rate": 1.9561928549563968e-05, | |
| "loss": 0.1617, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.07788989201628607, | |
| "grad_norm": 0.3950757086277008, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.1044, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.07930607187112763, | |
| "grad_norm": 2.2838950157165527, | |
| "learning_rate": 1.622048961921699e-05, | |
| "loss": 0.1557, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.0807222517259692, | |
| "grad_norm": 0.9748347401618958, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.1041, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.08213843158081076, | |
| "grad_norm": 0.6119588613510132, | |
| "learning_rate": 1.3136133159493802e-05, | |
| "loss": 0.093, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.08355461143565233, | |
| "grad_norm": 0.511449933052063, | |
| "learning_rate": 1.1697777844051105e-05, | |
| "loss": 0.1383, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.0849707912904939, | |
| "grad_norm": 1.2149078845977783, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 0.1075, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.08638697114533546, | |
| "grad_norm": 1.0309619903564453, | |
| "learning_rate": 9.042397785550405e-06, | |
| "loss": 0.0967, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.08780315100017702, | |
| "grad_norm": 2.3510637283325195, | |
| "learning_rate": 7.830427709355725e-06, | |
| "loss": 0.1512, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.08921933085501858, | |
| "grad_norm": 1.107035517692566, | |
| "learning_rate": 6.698729810778065e-06, | |
| "loss": 0.0723, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.09063551070986015, | |
| "grad_norm": 0.5546442866325378, | |
| "learning_rate": 5.649458341088915e-06, | |
| "loss": 0.1117, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.09205169056470172, | |
| "grad_norm": 1.5751676559448242, | |
| "learning_rate": 4.684610648167503e-06, | |
| "loss": 0.1086, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.09346787041954328, | |
| "grad_norm": 0.3522906005382538, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0987, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.09488405027438485, | |
| "grad_norm": 0.4917367398738861, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.1069, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.09630023012922641, | |
| "grad_norm": 1.3632891178131104, | |
| "learning_rate": 2.314152462588659e-06, | |
| "loss": 0.1204, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.09771640998406797, | |
| "grad_norm": 0.27073511481285095, | |
| "learning_rate": 1.70370868554659e-06, | |
| "loss": 0.0979, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.09913258983890955, | |
| "grad_norm": 0.3300623595714569, | |
| "learning_rate": 1.1851996440033319e-06, | |
| "loss": 0.0899, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.10054876969375111, | |
| "grad_norm": 2.2044923305511475, | |
| "learning_rate": 7.596123493895991e-07, | |
| "loss": 0.1263, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.10196494954859267, | |
| "grad_norm": 1.6999863386154175, | |
| "learning_rate": 4.277569313094809e-07, | |
| "loss": 0.1534, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.10338112940343423, | |
| "grad_norm": 0.6548103094100952, | |
| "learning_rate": 1.9026509541272275e-07, | |
| "loss": 0.1092, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.1047973092582758, | |
| "grad_norm": 0.9054380059242249, | |
| "learning_rate": 4.7588920907110094e-08, | |
| "loss": 0.1227, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.10621348911311737, | |
| "grad_norm": 0.3931428790092468, | |
| "learning_rate": 0.0, | |
| "loss": 0.1289, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.10621348911311737, | |
| "eval_loss": 0.10729946196079254, | |
| "eval_runtime": 153.0149, | |
| "eval_samples_per_second": 3.889, | |
| "eval_steps_per_second": 1.948, | |
| "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": 1.1302058350293811e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |