| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.975951903807615, | |
| "eval_steps": 500, | |
| "global_step": 124, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.16032064128256512, | |
| "grad_norm": 96.1263392680429, | |
| "learning_rate": 7.692307692307693e-05, | |
| "loss": 1.184, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.32064128256513025, | |
| "grad_norm": 109.03932941785152, | |
| "learning_rate": 0.00015384615384615385, | |
| "loss": 5.0239, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.48096192384769537, | |
| "grad_norm": 57.030962020859526, | |
| "learning_rate": 0.00019983983492623833, | |
| "loss": 4.0648, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.6412825651302605, | |
| "grad_norm": 22.869725355526718, | |
| "learning_rate": 0.0001980438647961327, | |
| "loss": 2.8094, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8016032064128257, | |
| "grad_norm": 6.985773379251052, | |
| "learning_rate": 0.00019428774454610843, | |
| "loss": 2.2094, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.9619238476953907, | |
| "grad_norm": 5.841783328387945, | |
| "learning_rate": 0.00018864656872260985, | |
| "loss": 1.3499, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.122244488977956, | |
| "grad_norm": 4.328203171108087, | |
| "learning_rate": 0.0001812331190023886, | |
| "loss": 1.2527, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.282565130260521, | |
| "grad_norm": 1.8271946123422904, | |
| "learning_rate": 0.00017219560939545246, | |
| "loss": 0.9903, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.4428857715430863, | |
| "grad_norm": 1.3376797882778457, | |
| "learning_rate": 0.00016171472306414554, | |
| "loss": 0.8082, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.6032064128256514, | |
| "grad_norm": 0.8625176200253305, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.7515, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.7635270541082164, | |
| "grad_norm": 0.9297452179429324, | |
| "learning_rate": 0.00013728564777803088, | |
| "loss": 0.6321, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.9238476953907817, | |
| "grad_norm": 1.1609492766676697, | |
| "learning_rate": 0.0001238258591423165, | |
| "loss": 0.5802, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.0841683366733466, | |
| "grad_norm": 0.6088051647121085, | |
| "learning_rate": 0.00010988973003642499, | |
| "loss": 0.4983, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.244488977955912, | |
| "grad_norm": 0.45623167145983035, | |
| "learning_rate": 9.57558796803852e-05, | |
| "loss": 0.4119, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.404809619238477, | |
| "grad_norm": 0.43383089288281307, | |
| "learning_rate": 8.170688025276134e-05, | |
| "loss": 0.3915, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.565130260521042, | |
| "grad_norm": 0.44731640463982364, | |
| "learning_rate": 6.802360754287547e-05, | |
| "loss": 0.3917, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.7254509018036073, | |
| "grad_norm": 0.4205909891049952, | |
| "learning_rate": 5.497962551823266e-05, | |
| "loss": 0.3886, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.8857715430861726, | |
| "grad_norm": 0.4168469096572953, | |
| "learning_rate": 4.283571707415214e-05, | |
| "loss": 0.3689, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.0460921843687374, | |
| "grad_norm": 0.6798047385077177, | |
| "learning_rate": 3.1834670310046734e-05, | |
| "loss": 0.3212, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 3.2064128256513027, | |
| "grad_norm": 0.3963204162500924, | |
| "learning_rate": 2.2196424568156073e-05, | |
| "loss": 0.1699, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.3667334669338675, | |
| "grad_norm": 0.38329008287670924, | |
| "learning_rate": 1.4113673277957395e-05, | |
| "loss": 0.1585, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 3.527054108216433, | |
| "grad_norm": 0.34785356511529864, | |
| "learning_rate": 7.74801151675314e-06, | |
| "loss": 0.1499, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.687374749498998, | |
| "grad_norm": 0.3409235398527789, | |
| "learning_rate": 3.226705306650113e-06, | |
| "loss": 0.1438, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 3.847695390781563, | |
| "grad_norm": 0.365385237116056, | |
| "learning_rate": 6.401472380297091e-07, | |
| "loss": 0.1525, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.975951903807615, | |
| "step": 124, | |
| "total_flos": 135489736671232.0, | |
| "train_loss": 1.0211431388893435, | |
| "train_runtime": 8462.237, | |
| "train_samples_per_second": 3.773, | |
| "train_steps_per_second": 0.015 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 124, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "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": 135489736671232.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |