| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5397382269599245, | |
| "eval_steps": 500, | |
| "global_step": 8000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03373363918499528, | |
| "grad_norm": 10.598428726196289, | |
| "learning_rate": 0.00018487499999999998, | |
| "loss": 4.9477, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.03373363918499528, | |
| "eval_loss": 1.7841953039169312, | |
| "eval_runtime": 187.924, | |
| "eval_samples_per_second": 37.366, | |
| "eval_steps_per_second": 0.585, | |
| "eval_wer": 0.939621839248462, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.06746727836999056, | |
| "grad_norm": 8.396936416625977, | |
| "learning_rate": 0.0002919583333333333, | |
| "loss": 1.035, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.06746727836999056, | |
| "eval_loss": 1.2386150360107422, | |
| "eval_runtime": 187.3652, | |
| "eval_samples_per_second": 37.478, | |
| "eval_steps_per_second": 0.587, | |
| "eval_wer": 0.7850622516908277, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.10120091755498584, | |
| "grad_norm": 7.086328506469727, | |
| "learning_rate": 0.000271125, | |
| "loss": 0.8162, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.10120091755498584, | |
| "eval_loss": 1.0026829242706299, | |
| "eval_runtime": 187.3863, | |
| "eval_samples_per_second": 37.473, | |
| "eval_steps_per_second": 0.587, | |
| "eval_wer": 0.7022017300256448, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.13493455673998112, | |
| "grad_norm": 8.218804359436035, | |
| "learning_rate": 0.0002502916666666666, | |
| "loss": 0.702, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.13493455673998112, | |
| "eval_loss": 0.9614182710647583, | |
| "eval_runtime": 188.0011, | |
| "eval_samples_per_second": 37.351, | |
| "eval_steps_per_second": 0.585, | |
| "eval_wer": 0.6608644813244927, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.1686681959249764, | |
| "grad_norm": 13.4183931350708, | |
| "learning_rate": 0.0002294583333333333, | |
| "loss": 0.6574, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.1686681959249764, | |
| "eval_loss": 0.8651247620582581, | |
| "eval_runtime": 189.0623, | |
| "eval_samples_per_second": 37.141, | |
| "eval_steps_per_second": 0.582, | |
| "eval_wer": 0.6165375569699305, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.20240183510997167, | |
| "grad_norm": 10.08790397644043, | |
| "learning_rate": 0.00020866666666666665, | |
| "loss": 0.602, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.20240183510997167, | |
| "eval_loss": 0.7873774766921997, | |
| "eval_runtime": 189.4729, | |
| "eval_samples_per_second": 37.061, | |
| "eval_steps_per_second": 0.581, | |
| "eval_wer": 0.5727952802987019, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.23613547429496695, | |
| "grad_norm": 24.115243911743164, | |
| "learning_rate": 0.0001878333333333333, | |
| "loss": 0.5547, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.23613547429496695, | |
| "eval_loss": 0.7803051471710205, | |
| "eval_runtime": 189.715, | |
| "eval_samples_per_second": 37.013, | |
| "eval_steps_per_second": 0.58, | |
| "eval_wer": 0.5766619274771124, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.26986911347996223, | |
| "grad_norm": 6.2811808586120605, | |
| "learning_rate": 0.00016699999999999997, | |
| "loss": 0.5325, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.26986911347996223, | |
| "eval_loss": 0.7444462776184082, | |
| "eval_runtime": 191.7775, | |
| "eval_samples_per_second": 36.615, | |
| "eval_steps_per_second": 0.574, | |
| "eval_wer": 0.5494359478600566, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.3036027526649575, | |
| "grad_norm": 7.9816060066223145, | |
| "learning_rate": 0.00014616666666666665, | |
| "loss": 0.5074, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.3036027526649575, | |
| "eval_loss": 0.6984341144561768, | |
| "eval_runtime": 191.1183, | |
| "eval_samples_per_second": 36.742, | |
| "eval_steps_per_second": 0.576, | |
| "eval_wer": 0.527963432944897, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.3373363918499528, | |
| "grad_norm": 12.77086353302002, | |
| "learning_rate": 0.00012533333333333334, | |
| "loss": 0.4755, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.3373363918499528, | |
| "eval_loss": 0.6562402248382568, | |
| "eval_runtime": 185.8168, | |
| "eval_samples_per_second": 37.79, | |
| "eval_steps_per_second": 0.592, | |
| "eval_wer": 0.49641903293958195, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.37107003103494807, | |
| "grad_norm": 6.413784980773926, | |
| "learning_rate": 0.00010449999999999999, | |
| "loss": 0.4582, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.37107003103494807, | |
| "eval_loss": 0.6308081746101379, | |
| "eval_runtime": 190.9299, | |
| "eval_samples_per_second": 36.778, | |
| "eval_steps_per_second": 0.576, | |
| "eval_wer": 0.4779361936778326, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.40480367021994335, | |
| "grad_norm": 6.88054084777832, | |
| "learning_rate": 8.370833333333333e-05, | |
| "loss": 0.4177, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.40480367021994335, | |
| "eval_loss": 0.6200515627861023, | |
| "eval_runtime": 188.1833, | |
| "eval_samples_per_second": 37.315, | |
| "eval_steps_per_second": 0.585, | |
| "eval_wer": 0.4672397985622982, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.4385373094049386, | |
| "grad_norm": 7.347259044647217, | |
| "learning_rate": 6.2875e-05, | |
| "loss": 0.3892, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.4385373094049386, | |
| "eval_loss": 0.5882463455200195, | |
| "eval_runtime": 186.5896, | |
| "eval_samples_per_second": 37.633, | |
| "eval_steps_per_second": 0.59, | |
| "eval_wer": 0.45023186595623116, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.4722709485899339, | |
| "grad_norm": 4.168457984924316, | |
| "learning_rate": 4.204166666666667e-05, | |
| "loss": 0.3694, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.4722709485899339, | |
| "eval_loss": 0.567690908908844, | |
| "eval_runtime": 186.8507, | |
| "eval_samples_per_second": 37.581, | |
| "eval_steps_per_second": 0.589, | |
| "eval_wer": 0.42950344809258695, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.5060045877749292, | |
| "grad_norm": 4.541583061218262, | |
| "learning_rate": 2.120833333333333e-05, | |
| "loss": 0.3437, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.5060045877749292, | |
| "eval_loss": 0.5473953485488892, | |
| "eval_runtime": 187.1269, | |
| "eval_samples_per_second": 37.525, | |
| "eval_steps_per_second": 0.588, | |
| "eval_wer": 0.4184615793459918, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.5397382269599245, | |
| "grad_norm": 8.075661659240723, | |
| "learning_rate": 3.7499999999999996e-07, | |
| "loss": 0.3481, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.5397382269599245, | |
| "eval_loss": 0.5434222221374512, | |
| "eval_runtime": 187.0779, | |
| "eval_samples_per_second": 37.535, | |
| "eval_steps_per_second": 0.588, | |
| "eval_wer": 0.4161495635073546, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.5397382269599245, | |
| "step": 8000, | |
| "total_flos": 8.953902963586946e+18, | |
| "train_loss": 0.8223040313720703, | |
| "train_runtime": 6199.5227, | |
| "train_samples_per_second": 10.323, | |
| "train_steps_per_second": 1.29 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 8000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "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": 8.953902963586946e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |