| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 40, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.025, | |
| "grad_norm": 2.1026898827355764, | |
| "learning_rate": 8.333333333333333e-07, | |
| "loss": 1.6621, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.6413522165034502, | |
| "learning_rate": 1.6666666666666667e-06, | |
| "loss": 1.7318, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.075, | |
| "grad_norm": 0.6085035314989437, | |
| "learning_rate": 2.5e-06, | |
| "loss": 1.6973, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.6097206000487838, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 1.7155, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 0.6181111292344054, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 1.7199, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.6095519945061052, | |
| "learning_rate": 5e-06, | |
| "loss": 1.7225, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.175, | |
| "grad_norm": 0.5993146111892984, | |
| "learning_rate": 5.833333333333334e-06, | |
| "loss": 1.6788, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.6168594833390011, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 1.6941, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.225, | |
| "grad_norm": 0.649510684690767, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 1.7237, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.6127108674374603, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 1.6575, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.275, | |
| "grad_norm": 0.6444847916730025, | |
| "learning_rate": 9.166666666666666e-06, | |
| "loss": 1.6477, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.6274944995465307, | |
| "learning_rate": 1e-05, | |
| "loss": 1.6443, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.325, | |
| "grad_norm": 0.6543139619848919, | |
| "learning_rate": 9.999836101895782e-06, | |
| "loss": 1.6202, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.8235053540937074, | |
| "learning_rate": 9.999344418328161e-06, | |
| "loss": 1.6082, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "grad_norm": 0.9775219557515483, | |
| "learning_rate": 9.99852498153154e-06, | |
| "loss": 1.5583, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 1.1030636654770976, | |
| "learning_rate": 9.997377845227577e-06, | |
| "loss": 1.5679, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.425, | |
| "grad_norm": 1.533566842443126, | |
| "learning_rate": 9.995903084621651e-06, | |
| "loss": 1.5359, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 1.718620417025083, | |
| "learning_rate": 9.994100796397954e-06, | |
| "loss": 1.4736, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.475, | |
| "grad_norm": 1.848832331677319, | |
| "learning_rate": 9.991971098713136e-06, | |
| "loss": 1.4416, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 1.1378030725614972, | |
| "learning_rate": 9.98951413118856e-06, | |
| "loss": 1.4385, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.525, | |
| "grad_norm": 0.7491289081750082, | |
| "learning_rate": 9.986730054901153e-06, | |
| "loss": 1.3793, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.6818788464479123, | |
| "learning_rate": 9.983619052372847e-06, | |
| "loss": 1.368, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.575, | |
| "grad_norm": 0.6341413849629475, | |
| "learning_rate": 9.98018132755861e-06, | |
| "loss": 1.3467, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.5923369379348106, | |
| "learning_rate": 9.97641710583307e-06, | |
| "loss": 1.3555, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 0.5340471860637729, | |
| "learning_rate": 9.972326633975754e-06, | |
| "loss": 1.349, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.4785970184408438, | |
| "learning_rate": 9.96791018015489e-06, | |
| "loss": 1.3337, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.675, | |
| "grad_norm": 0.4414536685738152, | |
| "learning_rate": 9.963168033909843e-06, | |
| "loss": 1.3334, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.4860929708817065, | |
| "learning_rate": 9.958100506132127e-06, | |
| "loss": 1.3159, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.725, | |
| "grad_norm": 0.45181979183711657, | |
| "learning_rate": 9.952707929045018e-06, | |
| "loss": 1.3231, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.4400284453917975, | |
| "learning_rate": 9.946990656181782e-06, | |
| "loss": 1.2993, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.775, | |
| "grad_norm": 0.38337739876412114, | |
| "learning_rate": 9.94094906236249e-06, | |
| "loss": 1.2952, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.37181863514378, | |
| "learning_rate": 9.934583543669454e-06, | |
| "loss": 1.3109, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.825, | |
| "grad_norm": 0.3401881416689356, | |
| "learning_rate": 9.927894517421252e-06, | |
| "loss": 1.3157, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.3239891568707606, | |
| "learning_rate": 9.920882422145372e-06, | |
| "loss": 1.34, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.875, | |
| "grad_norm": 0.3033475087643797, | |
| "learning_rate": 9.913547717549463e-06, | |
| "loss": 1.283, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.2888152197430936, | |
| "learning_rate": 9.905890884491196e-06, | |
| "loss": 1.2892, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.925, | |
| "grad_norm": 0.28458390089996766, | |
| "learning_rate": 9.897912424946741e-06, | |
| "loss": 1.2646, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.27339498391168426, | |
| "learning_rate": 9.889612861977855e-06, | |
| "loss": 1.2832, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.975, | |
| "grad_norm": 0.27940221076287475, | |
| "learning_rate": 9.88099273969759e-06, | |
| "loss": 1.2422, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.27720461090670556, | |
| "learning_rate": 9.872052623234632e-06, | |
| "loss": 1.2919, | |
| "step": 40 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 400, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 200.0, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 45589382299648.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |