| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0179640718562875, | |
| "eval_steps": 11, | |
| "global_step": 42, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.023952095808383235, | |
| "grad_norm": 17.82002830505371, | |
| "learning_rate": 5e-06, | |
| "loss": 1.8526, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.023952095808383235, | |
| "eval_loss": 0.46910959482192993, | |
| "eval_runtime": 0.6314, | |
| "eval_samples_per_second": 28.506, | |
| "eval_steps_per_second": 14.253, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.04790419161676647, | |
| "grad_norm": 17.064659118652344, | |
| "learning_rate": 1e-05, | |
| "loss": 1.812, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0718562874251497, | |
| "grad_norm": 16.225732803344727, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.7213, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.09580838323353294, | |
| "grad_norm": 18.819961547851562, | |
| "learning_rate": 2e-05, | |
| "loss": 1.9561, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.11976047904191617, | |
| "grad_norm": 16.86688804626465, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.7542, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.1437125748502994, | |
| "grad_norm": 14.079854965209961, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3857, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.16766467065868262, | |
| "grad_norm": 11.944511413574219, | |
| "learning_rate": 3.5e-05, | |
| "loss": 1.0198, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.19161676646706588, | |
| "grad_norm": 13.159642219543457, | |
| "learning_rate": 4e-05, | |
| "loss": 0.7081, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.2155688622754491, | |
| "grad_norm": 12.060245513916016, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.4168, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.23952095808383234, | |
| "grad_norm": 15.520999908447266, | |
| "learning_rate": 5e-05, | |
| "loss": 0.222, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2634730538922156, | |
| "grad_norm": 4.751951217651367, | |
| "learning_rate": 4.987961816680492e-05, | |
| "loss": 0.0575, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.2634730538922156, | |
| "eval_loss": 0.0022407574579119682, | |
| "eval_runtime": 0.6278, | |
| "eval_samples_per_second": 28.669, | |
| "eval_steps_per_second": 14.335, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.2874251497005988, | |
| "grad_norm": 0.9164900183677673, | |
| "learning_rate": 4.951963201008076e-05, | |
| "loss": 0.0086, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.31137724550898205, | |
| "grad_norm": 0.10736937820911407, | |
| "learning_rate": 4.892350839330522e-05, | |
| "loss": 0.002, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.33532934131736525, | |
| "grad_norm": 0.07489761710166931, | |
| "learning_rate": 4.8096988312782174e-05, | |
| "loss": 0.0014, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.3592814371257485, | |
| "grad_norm": 0.0928790420293808, | |
| "learning_rate": 4.7048031608708876e-05, | |
| "loss": 0.0015, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.38323353293413176, | |
| "grad_norm": 0.09091983735561371, | |
| "learning_rate": 4.5786740307563636e-05, | |
| "loss": 0.0012, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.40718562874251496, | |
| "grad_norm": 0.069945327937603, | |
| "learning_rate": 4.4325261334068426e-05, | |
| "loss": 0.001, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.4311377245508982, | |
| "grad_norm": 0.09426593780517578, | |
| "learning_rate": 4.267766952966369e-05, | |
| "loss": 0.0012, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.4550898203592814, | |
| "grad_norm": 0.03070351853966713, | |
| "learning_rate": 4.085983210409114e-05, | |
| "loss": 0.0008, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.47904191616766467, | |
| "grad_norm": 0.038840457797050476, | |
| "learning_rate": 3.888925582549006e-05, | |
| "loss": 0.0009, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.5029940119760479, | |
| "grad_norm": 0.026970243081450462, | |
| "learning_rate": 3.678491842064995e-05, | |
| "loss": 0.0007, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.5269461077844312, | |
| "grad_norm": 0.02901943400502205, | |
| "learning_rate": 3.456708580912725e-05, | |
| "loss": 0.0007, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.5269461077844312, | |
| "eval_loss": 0.00015069666551426053, | |
| "eval_runtime": 0.6489, | |
| "eval_samples_per_second": 27.738, | |
| "eval_steps_per_second": 13.869, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.5508982035928144, | |
| "grad_norm": 0.022013161331415176, | |
| "learning_rate": 3.225711693136156e-05, | |
| "loss": 0.0006, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.5748502994011976, | |
| "grad_norm": 0.019506849348545074, | |
| "learning_rate": 2.9877258050403212e-05, | |
| "loss": 0.0006, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.5988023952095808, | |
| "grad_norm": 0.013034597970545292, | |
| "learning_rate": 2.7450428508239024e-05, | |
| "loss": 0.0005, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.6227544910179641, | |
| "grad_norm": 0.015207488089799881, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0005, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.6467065868263473, | |
| "grad_norm": 0.020318390801548958, | |
| "learning_rate": 2.2549571491760986e-05, | |
| "loss": 0.0005, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.6706586826347305, | |
| "grad_norm": 0.011285997927188873, | |
| "learning_rate": 2.0122741949596797e-05, | |
| "loss": 0.0004, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.6946107784431138, | |
| "grad_norm": 0.01066031027585268, | |
| "learning_rate": 1.7742883068638447e-05, | |
| "loss": 0.0004, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.718562874251497, | |
| "grad_norm": 0.00946000311523676, | |
| "learning_rate": 1.5432914190872757e-05, | |
| "loss": 0.0004, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.7425149700598802, | |
| "grad_norm": 0.012651389464735985, | |
| "learning_rate": 1.3215081579350058e-05, | |
| "loss": 0.0004, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.7664670658682635, | |
| "grad_norm": 0.009206080809235573, | |
| "learning_rate": 1.1110744174509952e-05, | |
| "loss": 0.0004, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.7904191616766467, | |
| "grad_norm": 0.008822786621749401, | |
| "learning_rate": 9.140167895908867e-06, | |
| "loss": 0.0004, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.7904191616766467, | |
| "eval_loss": 0.00010356305574532598, | |
| "eval_runtime": 0.6264, | |
| "eval_samples_per_second": 28.734, | |
| "eval_steps_per_second": 14.367, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.8143712574850299, | |
| "grad_norm": 0.0093303257599473, | |
| "learning_rate": 7.3223304703363135e-06, | |
| "loss": 0.0004, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.8383233532934131, | |
| "grad_norm": 0.01624024100601673, | |
| "learning_rate": 5.674738665931575e-06, | |
| "loss": 0.0005, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.8622754491017964, | |
| "grad_norm": 0.010706230998039246, | |
| "learning_rate": 4.213259692436367e-06, | |
| "loss": 0.0004, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.8862275449101796, | |
| "grad_norm": 0.01006079837679863, | |
| "learning_rate": 2.9519683912911266e-06, | |
| "loss": 0.0004, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.9101796407185628, | |
| "grad_norm": 0.009781165048480034, | |
| "learning_rate": 1.9030116872178316e-06, | |
| "loss": 0.0004, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.9341317365269461, | |
| "grad_norm": 0.01016131229698658, | |
| "learning_rate": 1.0764916066947794e-06, | |
| "loss": 0.0004, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.9580838323353293, | |
| "grad_norm": 0.009532199241220951, | |
| "learning_rate": 4.803679899192392e-07, | |
| "loss": 0.0004, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.9820359281437125, | |
| "grad_norm": 0.010435133241117, | |
| "learning_rate": 1.2038183319507955e-07, | |
| "loss": 0.0004, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.0179640718562875, | |
| "grad_norm": 0.008128279820084572, | |
| "learning_rate": 0.0, | |
| "loss": 0.0003, | |
| "step": 42 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 42, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 11, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7361213100982272.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |