| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5807200929152149, | |
| "eval_steps": 100, | |
| "global_step": 3000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.019357336430507164, | |
| "eval_loss": 3.5493686199188232, | |
| "eval_runtime": 162.4972, | |
| "eval_samples_per_second": 34.807, | |
| "eval_steps_per_second": 4.351, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.03871467286101433, | |
| "eval_loss": 3.0426220893859863, | |
| "eval_runtime": 160.0022, | |
| "eval_samples_per_second": 35.35, | |
| "eval_steps_per_second": 4.419, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.05807200929152149, | |
| "eval_loss": 2.8965415954589844, | |
| "eval_runtime": 160.2187, | |
| "eval_samples_per_second": 35.302, | |
| "eval_steps_per_second": 4.413, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.07742934572202866, | |
| "eval_loss": 1.826250433921814, | |
| "eval_runtime": 161.1204, | |
| "eval_samples_per_second": 35.104, | |
| "eval_steps_per_second": 4.388, | |
| "eval_wer": 0.9828762176822712, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.09678668215253582, | |
| "grad_norm": 4.890473365783691, | |
| "learning_rate": 0.0002982, | |
| "loss": 3.9715, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09678668215253582, | |
| "eval_loss": 1.3860080242156982, | |
| "eval_runtime": 161.4467, | |
| "eval_samples_per_second": 35.033, | |
| "eval_steps_per_second": 4.379, | |
| "eval_wer": 0.8748856542183563, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.11614401858304298, | |
| "eval_loss": 1.308407187461853, | |
| "eval_runtime": 161.0068, | |
| "eval_samples_per_second": 35.129, | |
| "eval_steps_per_second": 4.391, | |
| "eval_wer": 0.8153456051098522, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.13550135501355012, | |
| "eval_loss": 1.0549893379211426, | |
| "eval_runtime": 161.8736, | |
| "eval_samples_per_second": 34.941, | |
| "eval_steps_per_second": 4.368, | |
| "eval_wer": 0.7336906806181894, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.1548586914440573, | |
| "eval_loss": 1.0011754035949707, | |
| "eval_runtime": 161.7881, | |
| "eval_samples_per_second": 34.959, | |
| "eval_steps_per_second": 4.37, | |
| "eval_wer": 0.7190062749755259, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.17421602787456447, | |
| "eval_loss": 0.913667619228363, | |
| "eval_runtime": 162.0369, | |
| "eval_samples_per_second": 34.906, | |
| "eval_steps_per_second": 4.363, | |
| "eval_wer": 0.6752258830704049, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.19357336430507163, | |
| "grad_norm": 3.201047658920288, | |
| "learning_rate": 0.0002406, | |
| "loss": 1.0155, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.19357336430507163, | |
| "eval_loss": 0.8486206531524658, | |
| "eval_runtime": 161.6295, | |
| "eval_samples_per_second": 34.994, | |
| "eval_steps_per_second": 4.374, | |
| "eval_wer": 0.6469483718765547, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2129307007355788, | |
| "eval_loss": 0.8534524440765381, | |
| "eval_runtime": 161.4191, | |
| "eval_samples_per_second": 35.039, | |
| "eval_steps_per_second": 4.38, | |
| "eval_wer": 0.6111761968191812, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.23228803716608595, | |
| "eval_loss": 0.8349705934524536, | |
| "eval_runtime": 162.6145, | |
| "eval_samples_per_second": 34.782, | |
| "eval_steps_per_second": 4.348, | |
| "eval_wer": 0.6192807048514708, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.2516453735965931, | |
| "eval_loss": 0.7680675983428955, | |
| "eval_runtime": 161.7995, | |
| "eval_samples_per_second": 34.957, | |
| "eval_steps_per_second": 4.37, | |
| "eval_wer": 0.5669785431143779, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.27100271002710025, | |
| "eval_loss": 0.7377049326896667, | |
| "eval_runtime": 162.5136, | |
| "eval_samples_per_second": 34.803, | |
| "eval_steps_per_second": 4.35, | |
| "eval_wer": 0.5559211054227985, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.29036004645760743, | |
| "grad_norm": 5.448112487792969, | |
| "learning_rate": 0.00018059999999999997, | |
| "loss": 0.7987, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.29036004645760743, | |
| "eval_loss": 0.7129804491996765, | |
| "eval_runtime": 162.7944, | |
| "eval_samples_per_second": 34.743, | |
| "eval_steps_per_second": 4.343, | |
| "eval_wer": 0.5437242220474715, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.3097173828881146, | |
| "eval_loss": 0.7039781808853149, | |
| "eval_runtime": 162.4512, | |
| "eval_samples_per_second": 34.817, | |
| "eval_steps_per_second": 4.352, | |
| "eval_wer": 0.5451846383463594, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.32907471931862176, | |
| "eval_loss": 0.6728500127792358, | |
| "eval_runtime": 166.2937, | |
| "eval_samples_per_second": 34.012, | |
| "eval_steps_per_second": 4.252, | |
| "eval_wer": 0.5050954085153504, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.34843205574912894, | |
| "eval_loss": 0.6646420359611511, | |
| "eval_runtime": 164.3307, | |
| "eval_samples_per_second": 34.418, | |
| "eval_steps_per_second": 4.302, | |
| "eval_wer": 0.511338286979827, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.3677893921796361, | |
| "eval_loss": 0.6530969142913818, | |
| "eval_runtime": 163.5666, | |
| "eval_samples_per_second": 34.579, | |
| "eval_steps_per_second": 4.322, | |
| "eval_wer": 0.49691065782927574, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.38714672861014326, | |
| "grad_norm": 2.6550886631011963, | |
| "learning_rate": 0.00012059999999999999, | |
| "loss": 0.6851, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.38714672861014326, | |
| "eval_loss": 0.6413969397544861, | |
| "eval_runtime": 162.7473, | |
| "eval_samples_per_second": 34.753, | |
| "eval_steps_per_second": 4.344, | |
| "eval_wer": 0.5037954775240326, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.4065040650406504, | |
| "eval_loss": 0.6108531355857849, | |
| "eval_runtime": 163.3073, | |
| "eval_samples_per_second": 34.634, | |
| "eval_steps_per_second": 4.329, | |
| "eval_wer": 0.467654186259248, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.4258614014711576, | |
| "eval_loss": 0.6034538745880127, | |
| "eval_runtime": 164.2548, | |
| "eval_samples_per_second": 34.434, | |
| "eval_steps_per_second": 4.304, | |
| "eval_wer": 0.46924299080419185, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.4452187379016647, | |
| "eval_loss": 0.5801683664321899, | |
| "eval_runtime": 163.2622, | |
| "eval_samples_per_second": 34.644, | |
| "eval_steps_per_second": 4.33, | |
| "eval_wer": 0.458955882588949, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.4645760743321719, | |
| "eval_loss": 0.571967363357544, | |
| "eval_runtime": 164.105, | |
| "eval_samples_per_second": 34.466, | |
| "eval_steps_per_second": 4.308, | |
| "eval_wer": 0.44545906822230424, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.48393341076267904, | |
| "grad_norm": 6.736985683441162, | |
| "learning_rate": 6.0599999999999996e-05, | |
| "loss": 0.5979, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.48393341076267904, | |
| "eval_loss": 0.569513201713562, | |
| "eval_runtime": 163.6068, | |
| "eval_samples_per_second": 34.571, | |
| "eval_steps_per_second": 4.321, | |
| "eval_wer": 0.4425542841552856, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.5032907471931862, | |
| "eval_loss": 0.5556703209877014, | |
| "eval_runtime": 163.3416, | |
| "eval_samples_per_second": 34.627, | |
| "eval_steps_per_second": 4.328, | |
| "eval_wer": 0.43513986294554735, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.5226480836236934, | |
| "eval_loss": 0.5499459505081177, | |
| "eval_runtime": 163.0587, | |
| "eval_samples_per_second": 34.687, | |
| "eval_steps_per_second": 4.336, | |
| "eval_wer": 0.4269711607902297, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.5420054200542005, | |
| "eval_loss": 0.5451160073280334, | |
| "eval_runtime": 163.5278, | |
| "eval_samples_per_second": 34.587, | |
| "eval_steps_per_second": 4.323, | |
| "eval_wer": 0.425815666575725, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.5613627564847077, | |
| "eval_loss": 0.5382982492446899, | |
| "eval_runtime": 163.1776, | |
| "eval_samples_per_second": 34.662, | |
| "eval_steps_per_second": 4.333, | |
| "eval_wer": 0.42172329123268765, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.5807200929152149, | |
| "grad_norm": 4.4167070388793945, | |
| "learning_rate": 6e-07, | |
| "loss": 0.5753, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.5807200929152149, | |
| "eval_loss": 0.5355480313301086, | |
| "eval_runtime": 164.5272, | |
| "eval_samples_per_second": 34.377, | |
| "eval_steps_per_second": 4.297, | |
| "eval_wer": 0.41856173067355684, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.5807200929152149, | |
| "step": 3000, | |
| "total_flos": 3.3874766991231493e+18, | |
| "train_loss": 1.2740035095214843, | |
| "train_runtime": 6100.931, | |
| "train_samples_per_second": 3.934, | |
| "train_steps_per_second": 0.492 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 400, | |
| "total_flos": 3.3874766991231493e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |