| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.4298480786416443, | |
| "eval_steps": 200, | |
| "global_step": 800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4292, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 4.994863481875841e-05, | |
| "loss": 0.2746, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_f1_score": 0.6564825695260478, | |
| "eval_label_f1": 0.8343125734430082, | |
| "eval_loss": 0.26023727655410767, | |
| "eval_runtime": 344.2386, | |
| "eval_samples_per_second": 2.905, | |
| "eval_steps_per_second": 0.363, | |
| "eval_wer": 0.10898676368139949, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 4.979475034558115e-05, | |
| "loss": 0.2568, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 4.9538978924776634e-05, | |
| "loss": 0.2481, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_f1_score": 0.6577916992952232, | |
| "eval_label_f1": 0.8347689898198903, | |
| "eval_loss": 0.246540829539299, | |
| "eval_runtime": 341.9234, | |
| "eval_samples_per_second": 2.925, | |
| "eval_steps_per_second": 0.366, | |
| "eval_wer": 0.10217509095131203, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 4.9182371575975736e-05, | |
| "loss": 0.2412, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 4.8726393675266716e-05, | |
| "loss": 0.2385, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_f1_score": 0.6684952978056427, | |
| "eval_label_f1": 0.8322884012539185, | |
| "eval_loss": 0.24104812741279602, | |
| "eval_runtime": 342.1463, | |
| "eval_samples_per_second": 2.923, | |
| "eval_steps_per_second": 0.365, | |
| "eval_wer": 0.10484557628299404, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 4.817291893365055e-05, | |
| "loss": 0.2325, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 4.752422169756048e-05, | |
| "loss": 0.2316, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "eval_f1_score": 0.6724477729601892, | |
| "eval_label_f1": 0.8316909735908553, | |
| "eval_loss": 0.23740312457084656, | |
| "eval_runtime": 344.4982, | |
| "eval_samples_per_second": 2.903, | |
| "eval_steps_per_second": 0.363, | |
| "eval_wer": 0.10221379363727842, | |
| "step": 800 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 200, | |
| "total_flos": 3.149300533694832e+19, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |