| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 8.936550491510276, |
| "eval_steps": 200, |
| "global_step": 5000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.18, |
| "learning_rate": 2e-05, |
| "loss": 1.1831, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.36, |
| "learning_rate": 4e-05, |
| "loss": 0.3562, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.36, |
| "eval_loss": 0.3264639973640442, |
| "eval_runtime": 194.3024, |
| "eval_samples_per_second": 5.147, |
| "eval_steps_per_second": 0.643, |
| "eval_wer": 0.19195022054190297, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.54, |
| "learning_rate": 6e-05, |
| "loss": 0.3149, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.71, |
| "learning_rate": 8e-05, |
| "loss": 0.3149, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.71, |
| "eval_loss": 0.313632994890213, |
| "eval_runtime": 190.5904, |
| "eval_samples_per_second": 5.247, |
| "eval_steps_per_second": 0.656, |
| "eval_wer": 0.18415248897290484, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.89, |
| "learning_rate": 0.0001, |
| "loss": 0.3164, |
| "step": 500 |
| }, |
| { |
| "epoch": 1.07, |
| "learning_rate": 9.987820251299122e-05, |
| "loss": 0.2778, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.07, |
| "eval_loss": 0.3204321563243866, |
| "eval_runtime": 190.0545, |
| "eval_samples_per_second": 5.262, |
| "eval_steps_per_second": 0.658, |
| "eval_wer": 0.1786389413988658, |
| "step": 600 |
| }, |
| { |
| "epoch": 1.25, |
| "learning_rate": 9.951340343707852e-05, |
| "loss": 0.2254, |
| "step": 700 |
| }, |
| { |
| "epoch": 1.43, |
| "learning_rate": 9.890738003669029e-05, |
| "loss": 0.2288, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.43, |
| "eval_loss": 0.31555524468421936, |
| "eval_runtime": 189.1302, |
| "eval_samples_per_second": 5.287, |
| "eval_steps_per_second": 0.661, |
| "eval_wer": 0.17166824196597352, |
| "step": 800 |
| }, |
| { |
| "epoch": 1.61, |
| "learning_rate": 9.806308479691595e-05, |
| "loss": 0.2295, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.79, |
| "learning_rate": 9.698463103929542e-05, |
| "loss": 0.2307, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.79, |
| "eval_loss": 0.30555686354637146, |
| "eval_runtime": 186.4985, |
| "eval_samples_per_second": 5.362, |
| "eval_steps_per_second": 0.67, |
| "eval_wer": 0.17080182734719596, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.97, |
| "learning_rate": 9.567727288213005e-05, |
| "loss": 0.2281, |
| "step": 1100 |
| }, |
| { |
| "epoch": 2.14, |
| "learning_rate": 9.414737964294636e-05, |
| "loss": 0.1482, |
| "step": 1200 |
| }, |
| { |
| "epoch": 2.14, |
| "eval_loss": 0.3138311207294464, |
| "eval_runtime": 188.367, |
| "eval_samples_per_second": 5.309, |
| "eval_steps_per_second": 0.664, |
| "eval_wer": 0.16816320100819154, |
| "step": 1200 |
| }, |
| { |
| "epoch": 2.32, |
| "learning_rate": 9.24024048078213e-05, |
| "loss": 0.1326, |
| "step": 1300 |
| }, |
| { |
| "epoch": 2.5, |
| "learning_rate": 9.045084971874738e-05, |
| "loss": 0.1368, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.5, |
| "eval_loss": 0.3136378824710846, |
| "eval_runtime": 188.5481, |
| "eval_samples_per_second": 5.304, |
| "eval_steps_per_second": 0.663, |
| "eval_wer": 0.16560333963453056, |
| "step": 1400 |
| }, |
| { |
| "epoch": 2.68, |
| "learning_rate": 8.83022221559489e-05, |
| "loss": 0.1415, |
| "step": 1500 |
| }, |
| { |
| "epoch": 2.86, |
| "learning_rate": 8.596699001693255e-05, |
| "loss": 0.1405, |
| "step": 1600 |
| }, |
| { |
| "epoch": 2.86, |
| "eval_loss": 0.30823495984077454, |
| "eval_runtime": 188.0037, |
| "eval_samples_per_second": 5.319, |
| "eval_steps_per_second": 0.665, |
| "eval_wer": 0.1617044738500315, |
| "step": 1600 |
| }, |
| { |
| "epoch": 3.04, |
| "learning_rate": 8.345653031794292e-05, |
| "loss": 0.1271, |
| "step": 1700 |
| }, |
| { |
| "epoch": 3.22, |
| "learning_rate": 8.07830737662829e-05, |
| "loss": 0.0639, |
| "step": 1800 |
| }, |
| { |
| "epoch": 3.22, |
| "eval_loss": 0.3200896680355072, |
| "eval_runtime": 188.8387, |
| "eval_samples_per_second": 5.296, |
| "eval_steps_per_second": 0.662, |
| "eval_wer": 0.1611925015752993, |
| "step": 1800 |
| }, |
| { |
| "epoch": 3.4, |
| "learning_rate": 7.795964517353735e-05, |
| "loss": 0.0644, |
| "step": 1900 |
| }, |
| { |
| "epoch": 3.57, |
| "learning_rate": 7.500000000000001e-05, |
| "loss": 0.0673, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.57, |
| "eval_loss": 0.3242465853691101, |
| "eval_runtime": 187.6919, |
| "eval_samples_per_second": 5.328, |
| "eval_steps_per_second": 0.666, |
| "eval_wer": 0.1611925015752993, |
| "step": 2000 |
| }, |
| { |
| "epoch": 3.75, |
| "learning_rate": 7.191855733945387e-05, |
| "loss": 0.0692, |
| "step": 2100 |
| }, |
| { |
| "epoch": 3.93, |
| "learning_rate": 6.873032967079561e-05, |
| "loss": 0.0688, |
| "step": 2200 |
| }, |
| { |
| "epoch": 3.93, |
| "eval_loss": 0.3234550952911377, |
| "eval_runtime": 188.2056, |
| "eval_samples_per_second": 5.313, |
| "eval_steps_per_second": 0.664, |
| "eval_wer": 0.15835696282293635, |
| "step": 2200 |
| }, |
| { |
| "epoch": 4.11, |
| "learning_rate": 6.545084971874738e-05, |
| "loss": 0.0424, |
| "step": 2300 |
| }, |
| { |
| "epoch": 4.29, |
| "learning_rate": 6.209609477998338e-05, |
| "loss": 0.0227, |
| "step": 2400 |
| }, |
| { |
| "epoch": 4.29, |
| "eval_loss": 0.3419637382030487, |
| "eval_runtime": 187.5691, |
| "eval_samples_per_second": 5.331, |
| "eval_steps_per_second": 0.666, |
| "eval_wer": 0.15575771896660365, |
| "step": 2400 |
| }, |
| { |
| "epoch": 4.47, |
| "learning_rate": 5.868240888334653e-05, |
| "loss": 0.0227, |
| "step": 2500 |
| }, |
| { |
| "epoch": 4.65, |
| "learning_rate": 5.522642316338268e-05, |
| "loss": 0.0232, |
| "step": 2600 |
| }, |
| { |
| "epoch": 4.65, |
| "eval_loss": 0.343047559261322, |
| "eval_runtime": 187.938, |
| "eval_samples_per_second": 5.321, |
| "eval_steps_per_second": 0.665, |
| "eval_wer": 0.15248897290485192, |
| "step": 2600 |
| }, |
| { |
| "epoch": 4.83, |
| "learning_rate": 5.174497483512506e-05, |
| "loss": 0.0235, |
| "step": 2700 |
| }, |
| { |
| "epoch": 5.0, |
| "learning_rate": 4.825502516487497e-05, |
| "loss": 0.0229, |
| "step": 2800 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 0.3450343608856201, |
| "eval_runtime": 187.1866, |
| "eval_samples_per_second": 5.342, |
| "eval_steps_per_second": 0.668, |
| "eval_wer": 0.15276465028355388, |
| "step": 2800 |
| }, |
| { |
| "epoch": 5.18, |
| "learning_rate": 4.477357683661734e-05, |
| "loss": 0.0071, |
| "step": 2900 |
| }, |
| { |
| "epoch": 5.36, |
| "learning_rate": 4.131759111665349e-05, |
| "loss": 0.0064, |
| "step": 3000 |
| }, |
| { |
| "epoch": 5.36, |
| "eval_loss": 0.36311542987823486, |
| "eval_runtime": 188.0361, |
| "eval_samples_per_second": 5.318, |
| "eval_steps_per_second": 0.665, |
| "eval_wer": 0.1498109640831758, |
| "step": 3000 |
| }, |
| { |
| "epoch": 5.54, |
| "learning_rate": 3.790390522001662e-05, |
| "loss": 0.0062, |
| "step": 3100 |
| }, |
| { |
| "epoch": 5.72, |
| "learning_rate": 3.4549150281252636e-05, |
| "loss": 0.0059, |
| "step": 3200 |
| }, |
| { |
| "epoch": 5.72, |
| "eval_loss": 0.36524486541748047, |
| "eval_runtime": 188.0737, |
| "eval_samples_per_second": 5.317, |
| "eval_steps_per_second": 0.665, |
| "eval_wer": 0.1482356647763075, |
| "step": 3200 |
| }, |
| { |
| "epoch": 5.9, |
| "learning_rate": 3.12696703292044e-05, |
| "loss": 0.0057, |
| "step": 3300 |
| }, |
| { |
| "epoch": 6.08, |
| "learning_rate": 2.8081442660546125e-05, |
| "loss": 0.0043, |
| "step": 3400 |
| }, |
| { |
| "epoch": 6.08, |
| "eval_loss": 0.3756215274333954, |
| "eval_runtime": 187.8897, |
| "eval_samples_per_second": 5.322, |
| "eval_steps_per_second": 0.665, |
| "eval_wer": 0.14815689981096408, |
| "step": 3400 |
| }, |
| { |
| "epoch": 6.26, |
| "learning_rate": 2.500000000000001e-05, |
| "loss": 0.0023, |
| "step": 3500 |
| }, |
| { |
| "epoch": 6.43, |
| "learning_rate": 2.2040354826462668e-05, |
| "loss": 0.0021, |
| "step": 3600 |
| }, |
| { |
| "epoch": 6.43, |
| "eval_loss": 0.3798007071018219, |
| "eval_runtime": 187.5918, |
| "eval_samples_per_second": 5.331, |
| "eval_steps_per_second": 0.666, |
| "eval_wer": 0.1476843100189036, |
| "step": 3600 |
| }, |
| { |
| "epoch": 6.61, |
| "learning_rate": 1.9216926233717085e-05, |
| "loss": 0.0021, |
| "step": 3700 |
| }, |
| { |
| "epoch": 6.79, |
| "learning_rate": 1.6543469682057106e-05, |
| "loss": 0.002, |
| "step": 3800 |
| }, |
| { |
| "epoch": 6.79, |
| "eval_loss": 0.3823593854904175, |
| "eval_runtime": 187.1697, |
| "eval_samples_per_second": 5.343, |
| "eval_steps_per_second": 0.668, |
| "eval_wer": 0.14839319470699433, |
| "step": 3800 |
| }, |
| { |
| "epoch": 6.97, |
| "learning_rate": 1.4033009983067452e-05, |
| "loss": 0.0019, |
| "step": 3900 |
| }, |
| { |
| "epoch": 7.15, |
| "learning_rate": 1.1697777844051105e-05, |
| "loss": 0.0014, |
| "step": 4000 |
| }, |
| { |
| "epoch": 7.15, |
| "eval_loss": 0.3876166343688965, |
| "eval_runtime": 186.567, |
| "eval_samples_per_second": 5.36, |
| "eval_steps_per_second": 0.67, |
| "eval_wer": 0.1471329552614997, |
| "step": 4000 |
| }, |
| { |
| "epoch": 7.33, |
| "learning_rate": 9.549150281252633e-06, |
| "loss": 0.0013, |
| "step": 4100 |
| }, |
| { |
| "epoch": 7.51, |
| "learning_rate": 7.597595192178702e-06, |
| "loss": 0.0013, |
| "step": 4200 |
| }, |
| { |
| "epoch": 7.51, |
| "eval_loss": 0.3900451064109802, |
| "eval_runtime": 187.3398, |
| "eval_samples_per_second": 5.338, |
| "eval_steps_per_second": 0.667, |
| "eval_wer": 0.14729048519218652, |
| "step": 4200 |
| }, |
| { |
| "epoch": 7.69, |
| "learning_rate": 5.852620357053651e-06, |
| "loss": 0.0013, |
| "step": 4300 |
| }, |
| { |
| "epoch": 7.86, |
| "learning_rate": 4.322727117869951e-06, |
| "loss": 0.0013, |
| "step": 4400 |
| }, |
| { |
| "epoch": 7.86, |
| "eval_loss": 0.39165398478507996, |
| "eval_runtime": 187.1352, |
| "eval_samples_per_second": 5.344, |
| "eval_steps_per_second": 0.668, |
| "eval_wer": 0.14610901071203528, |
| "step": 4400 |
| }, |
| { |
| "epoch": 8.04, |
| "learning_rate": 3.0153689607045845e-06, |
| "loss": 0.0013, |
| "step": 4500 |
| }, |
| { |
| "epoch": 8.22, |
| "learning_rate": 1.9369152030840556e-06, |
| "loss": 0.0012, |
| "step": 4600 |
| }, |
| { |
| "epoch": 8.22, |
| "eval_loss": 0.39290326833724976, |
| "eval_runtime": 187.7766, |
| "eval_samples_per_second": 5.325, |
| "eval_steps_per_second": 0.666, |
| "eval_wer": 0.1462271581600504, |
| "step": 4600 |
| }, |
| { |
| "epoch": 8.4, |
| "learning_rate": 1.0926199633097157e-06, |
| "loss": 0.0011, |
| "step": 4700 |
| }, |
| { |
| "epoch": 8.58, |
| "learning_rate": 4.865965629214819e-07, |
| "loss": 0.0012, |
| "step": 4800 |
| }, |
| { |
| "epoch": 8.58, |
| "eval_loss": 0.3932369351387024, |
| "eval_runtime": 187.8702, |
| "eval_samples_per_second": 5.323, |
| "eval_steps_per_second": 0.665, |
| "eval_wer": 0.14654221802142406, |
| "step": 4800 |
| }, |
| { |
| "epoch": 8.76, |
| "learning_rate": 1.2179748700879012e-07, |
| "loss": 0.0012, |
| "step": 4900 |
| }, |
| { |
| "epoch": 8.94, |
| "learning_rate": 0.0, |
| "loss": 0.0012, |
| "step": 5000 |
| }, |
| { |
| "epoch": 8.94, |
| "eval_loss": 0.3932703733444214, |
| "eval_runtime": 186.9555, |
| "eval_samples_per_second": 5.349, |
| "eval_steps_per_second": 0.669, |
| "eval_wer": 0.14642407057340895, |
| "step": 5000 |
| }, |
| { |
| "epoch": 8.94, |
| "step": 5000, |
| "total_flos": 1.8469234752159744e+20, |
| "train_loss": 0.1056724779933691, |
| "train_runtime": 20915.2691, |
| "train_samples_per_second": 30.6, |
| "train_steps_per_second": 0.239 |
| } |
| ], |
| "logging_steps": 100, |
| "max_steps": 5000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 9, |
| "save_steps": 200, |
| "total_flos": 1.8469234752159744e+20, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
