| { | |
| "best_metric": 6.602455701786633, | |
| "best_model_checkpoint": "checkpoint-130000", | |
| "epoch": 98.21512890735669, | |
| "eval_steps": 10000, | |
| "global_step": 130000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 7.56, | |
| "learning_rate": 3.7792894935752085e-05, | |
| "loss": 3.1181, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 7.56, | |
| "eval_cer": 12.22214052888268, | |
| "eval_loss": 0.39288055896759033, | |
| "eval_runtime": 80.6046, | |
| "eval_samples_per_second": 6.178, | |
| "eval_steps_per_second": 3.089, | |
| "eval_wer": 38.847117794486216, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 15.11, | |
| "learning_rate": 7.558578987150417e-05, | |
| "loss": 0.4086, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 15.11, | |
| "eval_cer": 9.646349533122564, | |
| "eval_loss": 0.3199174404144287, | |
| "eval_runtime": 78.777, | |
| "eval_samples_per_second": 6.322, | |
| "eval_steps_per_second": 3.161, | |
| "eval_wer": 30.367585630743527, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 22.67, | |
| "learning_rate": 9.665532879818595e-05, | |
| "loss": 0.3126, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 22.67, | |
| "eval_cer": 9.011592284881013, | |
| "eval_loss": 0.3147233724594116, | |
| "eval_runtime": 77.284, | |
| "eval_samples_per_second": 6.444, | |
| "eval_steps_per_second": 3.222, | |
| "eval_wer": 28.390420495683657, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 30.22, | |
| "learning_rate": 8.720710506424793e-05, | |
| "loss": 0.2509, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 30.22, | |
| "eval_cer": 8.134205818199643, | |
| "eval_loss": 0.3039480447769165, | |
| "eval_runtime": 68.0983, | |
| "eval_samples_per_second": 7.313, | |
| "eval_steps_per_second": 3.656, | |
| "eval_wer": 26.427179058758004, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 37.78, | |
| "learning_rate": 7.77588813303099e-05, | |
| "loss": 0.2084, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 37.78, | |
| "eval_cer": 7.7028649854177385, | |
| "eval_loss": 0.2937542498111725, | |
| "eval_runtime": 66.6192, | |
| "eval_samples_per_second": 7.475, | |
| "eval_steps_per_second": 3.738, | |
| "eval_wer": 25.382901698691175, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 45.33, | |
| "learning_rate": 6.831065759637189e-05, | |
| "loss": 0.1794, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 45.33, | |
| "eval_cer": 7.658750582065044, | |
| "eval_loss": 0.3241848349571228, | |
| "eval_runtime": 65.9465, | |
| "eval_samples_per_second": 7.552, | |
| "eval_steps_per_second": 3.776, | |
| "eval_wer": 24.770258980785297, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 52.89, | |
| "learning_rate": 5.886243386243386e-05, | |
| "loss": 0.1566, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 52.89, | |
| "eval_cer": 7.2984829546847045, | |
| "eval_loss": 0.33441099524497986, | |
| "eval_runtime": 66.4973, | |
| "eval_samples_per_second": 7.489, | |
| "eval_steps_per_second": 3.745, | |
| "eval_wer": 24.33862433862434, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 60.44, | |
| "learning_rate": 4.9414210128495846e-05, | |
| "loss": 0.1381, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 60.44, | |
| "eval_cer": 7.394064161948877, | |
| "eval_loss": 0.3713204860687256, | |
| "eval_runtime": 66.5533, | |
| "eval_samples_per_second": 7.483, | |
| "eval_steps_per_second": 3.741, | |
| "eval_wer": 23.62851573377889, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "learning_rate": 3.9965986394557825e-05, | |
| "loss": 0.1227, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_cer": 7.109771340342622, | |
| "eval_loss": 0.3827340006828308, | |
| "eval_runtime": 66.2381, | |
| "eval_samples_per_second": 7.518, | |
| "eval_steps_per_second": 3.759, | |
| "eval_wer": 22.946254525201894, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 75.55, | |
| "learning_rate": 3.0517762660619804e-05, | |
| "loss": 0.1097, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 75.55, | |
| "eval_cer": 7.124476141460187, | |
| "eval_loss": 0.415243923664093, | |
| "eval_runtime": 149.6844, | |
| "eval_samples_per_second": 3.327, | |
| "eval_steps_per_second": 1.663, | |
| "eval_wer": 22.988025619604567, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 83.11, | |
| "learning_rate": 2.1069538926681782e-05, | |
| "loss": 0.0988, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 83.11, | |
| "eval_cer": 6.854888120971498, | |
| "eval_loss": 0.4487506151199341, | |
| "eval_runtime": 66.6595, | |
| "eval_samples_per_second": 7.471, | |
| "eval_steps_per_second": 3.735, | |
| "eval_wer": 22.570314675577833, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 90.66, | |
| "learning_rate": 1.1621315192743764e-05, | |
| "loss": 0.0896, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 90.66, | |
| "eval_cer": 6.808322917432541, | |
| "eval_loss": 0.44282594323158264, | |
| "eval_runtime": 66.7552, | |
| "eval_samples_per_second": 7.46, | |
| "eval_steps_per_second": 3.73, | |
| "eval_wer": 21.99944305207463, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 98.22, | |
| "learning_rate": 2.1730914588057445e-06, | |
| "loss": 0.0823, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 98.22, | |
| "eval_cer": 6.602455701786633, | |
| "eval_loss": 0.4513276517391205, | |
| "eval_runtime": 66.6171, | |
| "eval_samples_per_second": 7.476, | |
| "eval_steps_per_second": 3.738, | |
| "eval_wer": 21.74881648565859, | |
| "step": 130000 | |
| } | |
| ], | |
| "logging_steps": 10000, | |
| "max_steps": 132300, | |
| "num_train_epochs": 100, | |
| "save_steps": 10000, | |
| "total_flos": 4.063479904391249e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |