| { | |
| "best_metric": 82.73972602739727, | |
| "best_model_checkpoint": "/home/rcj2772/data_sets/Seneca/output_dir/checkpoint-100", | |
| "epoch": 62.008, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 9.979166666666668e-06, | |
| "loss": 4.0703, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 9.458333333333334e-06, | |
| "loss": 0.9051, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "learning_rate": 8.9375e-06, | |
| "loss": 0.0795, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "learning_rate": 8.416666666666667e-06, | |
| "loss": 0.009, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "eval_loss": 2.0225086212158203, | |
| "eval_runtime": 26.0259, | |
| "eval_samples_per_second": 0.999, | |
| "eval_steps_per_second": 0.269, | |
| "eval_wer": 82.73972602739727, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 15.01, | |
| "learning_rate": 7.895833333333333e-06, | |
| "loss": 0.0015, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 18.01, | |
| "learning_rate": 7.375000000000001e-06, | |
| "loss": 0.0007, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 21.01, | |
| "learning_rate": 6.854166666666667e-06, | |
| "loss": 0.0005, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 24.02, | |
| "learning_rate": 6.333333333333333e-06, | |
| "loss": 0.0004, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 24.02, | |
| "eval_loss": 2.1611859798431396, | |
| "eval_runtime": 19.928, | |
| "eval_samples_per_second": 1.305, | |
| "eval_steps_per_second": 0.351, | |
| "eval_wer": 86.02739726027397, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "learning_rate": 5.812500000000001e-06, | |
| "loss": 0.0003, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "learning_rate": 5.291666666666667e-06, | |
| "loss": 0.0003, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 34.01, | |
| "learning_rate": 4.770833333333334e-06, | |
| "loss": 0.0003, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 37.01, | |
| "learning_rate": 4.25e-06, | |
| "loss": 0.0002, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 37.01, | |
| "eval_loss": 2.2043375968933105, | |
| "eval_runtime": 20.0279, | |
| "eval_samples_per_second": 1.298, | |
| "eval_steps_per_second": 0.35, | |
| "eval_wer": 85.2054794520548, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 40.01, | |
| "learning_rate": 3.7291666666666672e-06, | |
| "loss": 0.0002, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 43.01, | |
| "learning_rate": 3.2083333333333337e-06, | |
| "loss": 0.0002, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 46.01, | |
| "learning_rate": 2.6875e-06, | |
| "loss": 0.0002, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 49.02, | |
| "learning_rate": 2.166666666666667e-06, | |
| "loss": 0.0002, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 49.02, | |
| "eval_loss": 2.2259132862091064, | |
| "eval_runtime": 19.9281, | |
| "eval_samples_per_second": 1.305, | |
| "eval_steps_per_second": 0.351, | |
| "eval_wer": 85.47945205479452, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "learning_rate": 1.6458333333333334e-06, | |
| "loss": 0.0002, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "learning_rate": 1.125e-06, | |
| "loss": 0.0002, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 59.01, | |
| "learning_rate": 6.041666666666667e-07, | |
| "loss": 0.0002, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 62.01, | |
| "learning_rate": 8.333333333333334e-08, | |
| "loss": 0.0002, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 62.01, | |
| "eval_loss": 2.232715129852295, | |
| "eval_runtime": 20.22, | |
| "eval_samples_per_second": 1.286, | |
| "eval_steps_per_second": 0.346, | |
| "eval_wer": 85.2054794520548, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 62.01, | |
| "step": 500, | |
| "total_flos": 5.234938970541916e+17, | |
| "train_loss": 0.2534824407529086, | |
| "train_runtime": 588.6372, | |
| "train_samples_per_second": 3.398, | |
| "train_steps_per_second": 0.849 | |
| } | |
| ], | |
| "max_steps": 500, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 5.234938970541916e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |