| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.3, | |
| "eval_steps": 500, | |
| "global_step": 1500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.9600000000000003e-06, | |
| "loss": 4.1002, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3.96e-06, | |
| "loss": 3.4625, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 5.9600000000000005e-06, | |
| "loss": 2.4907, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 7.960000000000002e-06, | |
| "loss": 1.5881, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.960000000000001e-06, | |
| "loss": 1.1583, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_combined_wer": 0.3216761184625079, | |
| "eval_f1_score": 0.07455306200076076, | |
| "eval_label_f1": 0.1414986686953214, | |
| "eval_loss": 1.0361335277557373, | |
| "eval_runtime": 337.7385, | |
| "eval_samples_per_second": 2.961, | |
| "eval_steps_per_second": 0.047, | |
| "eval_wer": 0.20672983948946044, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.98830238119205e-06, | |
| "loss": 0.9384, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 9.952307128483257e-06, | |
| "loss": 0.7396, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.892184733248666e-06, | |
| "loss": 0.4787, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 9.808228105754378e-06, | |
| "loss": 0.436, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 9.700846274250252e-06, | |
| "loss": 0.4069, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_combined_wer": 0.22034499054820417, | |
| "eval_f1_score": 0.4222737819025522, | |
| "eval_label_f1": 0.5939675174013921, | |
| "eval_loss": 0.4110757112503052, | |
| "eval_runtime": 259.1805, | |
| "eval_samples_per_second": 3.858, | |
| "eval_steps_per_second": 0.062, | |
| "eval_wer": 0.12349642235544382, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 9.570562392225395e-06, | |
| "loss": 0.4044, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 9.418011189656942e-06, | |
| "loss": 0.3882, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 9.24393588066941e-06, | |
| "loss": 0.3855, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 9.0491845426702e-06, | |
| "loss": 0.3774, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 8.834705984601708e-06, | |
| "loss": 0.3708, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_combined_wer": 0.2200693131695022, | |
| "eval_f1_score": 0.4609408718170047, | |
| "eval_label_f1": 0.6266724212343547, | |
| "eval_loss": 0.3768249750137329, | |
| "eval_runtime": 253.1584, | |
| "eval_samples_per_second": 3.95, | |
| "eval_steps_per_second": 0.063, | |
| "eval_wer": 0.12949139431444595, | |
| "step": 1500 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 5.593838911488e+19, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |