| { | |
| "best_metric": 172.4755700325733, | |
| "best_model_checkpoint": "./temp/checkpoint-70", | |
| "epoch": 13.952380952380953, | |
| "eval_steps": 10, | |
| "global_step": 210, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.6349206349206349, | |
| "eval_cer": 186.56571539506893, | |
| "eval_loss": 9.826007843017578, | |
| "eval_runtime": 21.7924, | |
| "eval_samples_per_second": 2.294, | |
| "eval_steps_per_second": 0.184, | |
| "eval_wer": 332.2475570032573, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.3174603174603174, | |
| "eval_cer": 186.54630168899243, | |
| "eval_loss": 8.942676544189453, | |
| "eval_runtime": 20.9166, | |
| "eval_samples_per_second": 2.39, | |
| "eval_steps_per_second": 0.191, | |
| "eval_wer": 332.5732899022801, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.9523809523809523, | |
| "eval_cer": 179.30498932246167, | |
| "eval_loss": 7.41646671295166, | |
| "eval_runtime": 23.189, | |
| "eval_samples_per_second": 2.156, | |
| "eval_steps_per_second": 0.172, | |
| "eval_wer": 329.64169381107496, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.571428571428571, | |
| "eval_cer": 173.6167734420501, | |
| "eval_loss": 4.776045322418213, | |
| "eval_runtime": 23.185, | |
| "eval_samples_per_second": 2.157, | |
| "eval_steps_per_second": 0.173, | |
| "eval_wer": 313.51791530944627, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.1904761904761907, | |
| "grad_norm": 63047348.0, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 6.244, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.1904761904761907, | |
| "eval_cer": 173.05377596583187, | |
| "eval_loss": 1.3689286708831787, | |
| "eval_runtime": 20.2703, | |
| "eval_samples_per_second": 2.467, | |
| "eval_steps_per_second": 0.197, | |
| "eval_wer": 295.114006514658, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.825396825396825, | |
| "eval_cer": 119.18074160357213, | |
| "eval_loss": 0.5696829557418823, | |
| "eval_runtime": 19.3264, | |
| "eval_samples_per_second": 2.587, | |
| "eval_steps_per_second": 0.207, | |
| "eval_wer": 184.85342019543972, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.444444444444445, | |
| "eval_cer": 113.70607649000195, | |
| "eval_loss": 0.4699205756187439, | |
| "eval_runtime": 15.0547, | |
| "eval_samples_per_second": 3.321, | |
| "eval_steps_per_second": 0.266, | |
| "eval_wer": 172.4755700325733, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 5.317460317460317, | |
| "eval_cer": 126.20850320326151, | |
| "eval_loss": 0.41475537419319153, | |
| "eval_runtime": 23.4851, | |
| "eval_samples_per_second": 2.129, | |
| "eval_steps_per_second": 0.17, | |
| "eval_wer": 175.2442996742671, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 5.9523809523809526, | |
| "eval_cer": 127.43156668608037, | |
| "eval_loss": 0.37485015392303467, | |
| "eval_runtime": 19.6066, | |
| "eval_samples_per_second": 2.55, | |
| "eval_steps_per_second": 0.204, | |
| "eval_wer": 230.94462540716614, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 6.571428571428571, | |
| "grad_norm": 471759.03125, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.3668, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.571428571428571, | |
| "eval_cer": 146.02989710735778, | |
| "eval_loss": 0.3358791768550873, | |
| "eval_runtime": 25.2622, | |
| "eval_samples_per_second": 1.979, | |
| "eval_steps_per_second": 0.158, | |
| "eval_wer": 242.01954397394138, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 7.317460317460317, | |
| "eval_cer": 217.5305765870705, | |
| "eval_loss": 0.30219942331314087, | |
| "eval_runtime": 24.1771, | |
| "eval_samples_per_second": 2.068, | |
| "eval_steps_per_second": 0.165, | |
| "eval_wer": 297.0684039087948, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 7.9523809523809526, | |
| "eval_cer": 280.9163269268103, | |
| "eval_loss": 0.2763884663581848, | |
| "eval_runtime": 28.0408, | |
| "eval_samples_per_second": 1.783, | |
| "eval_steps_per_second": 0.143, | |
| "eval_wer": 312.5407166123778, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 8.634920634920634, | |
| "eval_cer": 232.82857697534462, | |
| "eval_loss": 0.25809770822525024, | |
| "eval_runtime": 26.074, | |
| "eval_samples_per_second": 1.918, | |
| "eval_steps_per_second": 0.153, | |
| "eval_wer": 364.98371335504885, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 9.317460317460318, | |
| "eval_cer": 247.01999611725878, | |
| "eval_loss": 0.24312858283519745, | |
| "eval_runtime": 25.4471, | |
| "eval_samples_per_second": 1.965, | |
| "eval_steps_per_second": 0.157, | |
| "eval_wer": 348.04560260586317, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 9.952380952380953, | |
| "grad_norm": 268960.5625, | |
| "learning_rate": 3e-06, | |
| "loss": 0.2195, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 9.952380952380953, | |
| "eval_cer": 282.1588041157057, | |
| "eval_loss": 0.23099038004875183, | |
| "eval_runtime": 26.2109, | |
| "eval_samples_per_second": 1.908, | |
| "eval_steps_per_second": 0.153, | |
| "eval_wer": 503.4201954397394, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 10.634920634920634, | |
| "eval_cer": 274.8204232187925, | |
| "eval_loss": 0.22056226432323456, | |
| "eval_runtime": 26.4315, | |
| "eval_samples_per_second": 1.892, | |
| "eval_steps_per_second": 0.151, | |
| "eval_wer": 514.1693811074919, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 11.317460317460318, | |
| "eval_cer": 289.0118423607066, | |
| "eval_loss": 0.2121584266424179, | |
| "eval_runtime": 27.1917, | |
| "eval_samples_per_second": 1.839, | |
| "eval_steps_per_second": 0.147, | |
| "eval_wer": 520.3583061889251, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 11.952380952380953, | |
| "eval_cer": 230.24655406717142, | |
| "eval_loss": 0.20317548513412476, | |
| "eval_runtime": 26.2613, | |
| "eval_samples_per_second": 1.904, | |
| "eval_steps_per_second": 0.152, | |
| "eval_wer": 469.21824104234526, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 12.634920634920634, | |
| "eval_cer": 193.67113181906427, | |
| "eval_loss": 0.1965637505054474, | |
| "eval_runtime": 27.8147, | |
| "eval_samples_per_second": 1.798, | |
| "eval_steps_per_second": 0.144, | |
| "eval_wer": 419.86970684039085, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 13.317460317460318, | |
| "grad_norm": 230837.75, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.166, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 13.317460317460318, | |
| "eval_cer": 277.96544360318387, | |
| "eval_loss": 0.1915261596441269, | |
| "eval_runtime": 27.6252, | |
| "eval_samples_per_second": 1.81, | |
| "eval_steps_per_second": 0.145, | |
| "eval_wer": 570.1954397394137, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 13.952380952380953, | |
| "eval_cer": 296.81615220345566, | |
| "eval_loss": 0.1847289502620697, | |
| "eval_runtime": 26.8778, | |
| "eval_samples_per_second": 1.86, | |
| "eval_steps_per_second": 0.149, | |
| "eval_wer": 619.8697068403908, | |
| "step": 210 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 210, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 14, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.2851440304128e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |