| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.1475104469860361, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0999, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.17336762271557374, | |
| "learning_rate": 4e-05, | |
| "loss": 0.1105, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.1625647711677745, | |
| "learning_rate": 6e-05, | |
| "loss": 0.1518, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.2835973362154201, | |
| "learning_rate": 8e-05, | |
| "loss": 0.412, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.102899700816259, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1253, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.10525052737112932, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.1777, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.0841421968850327, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.1138, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.21097849799520993, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.3093, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.13129190430279525, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.1067, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.22029292639589196, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.2229, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.2605739248411656, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.3048, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.09516786744052268, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.0444, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.11035350979508174, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.0919, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.1860366786152161, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.1798, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.2051102247541834, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.1682, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.2120620056755024, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.2017, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.12492795843072409, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.0715, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.09372961310287645, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.063, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.09404028728202063, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 0.1209, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.11155850164626473, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.048, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.14193160043170325, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.1016, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.12622860771717062, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 0.0836, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.19822006343379234, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.1943, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.17907254526794228, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.1258, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.35070363993403475, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.2436, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.1445501148700714, | |
| "eval_runtime": 0.4011, | |
| "eval_samples_per_second": 2.493, | |
| "eval_steps_per_second": 2.493, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 0.331374058760647, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 0.1818, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 0.10212151784996519, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.0545, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 0.12127508733134722, | |
| "learning_rate": 4.825502516487497e-05, | |
| "loss": 0.0662, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 0.10472782126959304, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 0.0712, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.09739912535877197, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.0716, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 0.07818016052332849, | |
| "learning_rate": 3.790390522001662e-05, | |
| "loss": 0.0428, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 0.1309525888712825, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.0632, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 0.05268163632164765, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.0308, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.3599999999999999, | |
| "grad_norm": 0.07825640200592207, | |
| "learning_rate": 2.8081442660546125e-05, | |
| "loss": 0.0439, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.0654461820686998, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.0337, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 0.11019625858497928, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.0732, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 0.0589572493253463, | |
| "learning_rate": 1.9216926233717085e-05, | |
| "loss": 0.0342, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 0.06589353609148312, | |
| "learning_rate": 1.6543469682057106e-05, | |
| "loss": 0.0325, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 0.14358958467272426, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.0941, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.10816135111790075, | |
| "learning_rate": 1.1697777844051105e-05, | |
| "loss": 0.0577, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.6400000000000001, | |
| "grad_norm": 0.06366691648082863, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.0333, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.6800000000000002, | |
| "grad_norm": 0.15879304612419673, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 0.0666, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 0.0765495019619371, | |
| "learning_rate": 5.852620357053651e-06, | |
| "loss": 0.029, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 0.1515518653937463, | |
| "learning_rate": 4.322727117869951e-06, | |
| "loss": 0.1331, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 0.08525038364295008, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.0489, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.8399999999999999, | |
| "grad_norm": 0.09069389566818667, | |
| "learning_rate": 1.9369152030840556e-06, | |
| "loss": 0.0417, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 0.15581325970621174, | |
| "learning_rate": 1.0926199633097157e-06, | |
| "loss": 0.1167, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.07551809989198507, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 0.0324, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 0.06320999636101195, | |
| "learning_rate": 1.2179748700879012e-07, | |
| "loss": 0.0494, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.13783812798955924, | |
| "learning_rate": 0.0, | |
| "loss": 0.12, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.14731274545192719, | |
| "eval_runtime": 1.4061, | |
| "eval_samples_per_second": 0.711, | |
| "eval_steps_per_second": 0.711, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 50, | |
| "total_flos": 643887562752.0, | |
| "train_loss": 0.10991490356624127, | |
| "train_runtime": 185.8623, | |
| "train_samples_per_second": 0.538, | |
| "train_steps_per_second": 0.269 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "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": 643887562752.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |