| { | |
| "best_metric": 82.71868753015785, | |
| "best_model_checkpoint": "/scratch/p310333/whisper-small-dialect_all_seed168/checkpoint-250", | |
| "epoch": 0.03017137340091721, | |
| "eval_steps": 250, | |
| "global_step": 250, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.003017137340091721, | |
| "grad_norm": 20.100784301757812, | |
| "learning_rate": 5.000000000000001e-07, | |
| "loss": 1.2639, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.006034274680183442, | |
| "grad_norm": 24.114206314086914, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 1.383, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.009051412020275163, | |
| "grad_norm": 18.229045867919922, | |
| "learning_rate": 1.5e-06, | |
| "loss": 1.2886, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.012068549360366883, | |
| "grad_norm": 21.113380432128906, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 1.2495, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.015085686700458605, | |
| "grad_norm": 19.328229904174805, | |
| "learning_rate": 2.5e-06, | |
| "loss": 1.2043, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.018102824040550327, | |
| "grad_norm": 25.50276756286621, | |
| "learning_rate": 3e-06, | |
| "loss": 1.273, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.021119961380642045, | |
| "grad_norm": 18.037517547607422, | |
| "learning_rate": 3.5e-06, | |
| "loss": 1.2175, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.024137098720733767, | |
| "grad_norm": 22.56538963317871, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.1856, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.02715423606082549, | |
| "grad_norm": 16.57452392578125, | |
| "learning_rate": 4.5e-06, | |
| "loss": 1.1098, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.03017137340091721, | |
| "grad_norm": 19.666919708251953, | |
| "learning_rate": 5e-06, | |
| "loss": 1.0496, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.03017137340091721, | |
| "eval_cer": 53.72986812504118, | |
| "eval_loss": 1.3066076040267944, | |
| "eval_runtime": 3968.9492, | |
| "eval_samples_per_second": 4.175, | |
| "eval_steps_per_second": 0.522, | |
| "eval_wer": 82.71868753015785, | |
| "step": 250 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 250, | |
| "total_flos": 5.7717080064e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |