| { | |
| "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": 0.0467, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 4.994863481875841e-05, | |
| "loss": 0.0433, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_f1_score": 0.6251468860164512, | |
| "eval_label_f1": 0.8319623971797884, | |
| "eval_loss": 0.05231842026114464, | |
| "eval_runtime": 457.2641, | |
| "eval_samples_per_second": 2.187, | |
| "eval_steps_per_second": 0.547, | |
| "eval_wer": 0.10434244136543076, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 4.979475034558115e-05, | |
| "loss": 0.0402, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 4.9538978924776634e-05, | |
| "loss": 0.0391, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_f1_score": 0.6206896551724138, | |
| "eval_label_f1": 0.8346394984326019, | |
| "eval_loss": 0.050447478890419006, | |
| "eval_runtime": 457.7184, | |
| "eval_samples_per_second": 2.185, | |
| "eval_steps_per_second": 0.546, | |
| "eval_wer": 0.10472946822509482, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 4.9182371575975736e-05, | |
| "loss": 0.037, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 4.8726393675266716e-05, | |
| "loss": 0.0381, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_f1_score": 0.6142467886337096, | |
| "eval_label_f1": 0.8322304398598676, | |
| "eval_loss": 0.049576789140701294, | |
| "eval_runtime": 458.238, | |
| "eval_samples_per_second": 2.182, | |
| "eval_steps_per_second": 0.546, | |
| "eval_wer": 0.10654849446551591, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 4.817291893365055e-05, | |
| "loss": 0.037, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 4.752422169756048e-05, | |
| "loss": 0.0374, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "eval_f1_score": 0.6158081006685017, | |
| "eval_label_f1": 0.8360204482894219, | |
| "eval_loss": 0.04838084056973457, | |
| "eval_runtime": 456.7741, | |
| "eval_samples_per_second": 2.189, | |
| "eval_steps_per_second": 0.547, | |
| "eval_wer": 0.10705162938307919, | |
| "step": 800 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 200, | |
| "total_flos": 3.126858539481072e+19, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |