| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 15.0, | |
| "eval_steps": 100, | |
| "global_step": 5640, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.26595744680851063, | |
| "eval_loss": 17.797901153564453, | |
| "eval_runtime": 208.2309, | |
| "eval_samples_per_second": 23.825, | |
| "eval_steps_per_second": 2.982, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5319148936170213, | |
| "eval_loss": 5.018890380859375, | |
| "eval_runtime": 206.5228, | |
| "eval_samples_per_second": 24.022, | |
| "eval_steps_per_second": 3.007, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7978723404255319, | |
| "eval_loss": 4.124919891357422, | |
| "eval_runtime": 206.6954, | |
| "eval_samples_per_second": 24.001, | |
| "eval_steps_per_second": 3.004, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.0638297872340425, | |
| "eval_loss": 4.171339511871338, | |
| "eval_runtime": 206.9498, | |
| "eval_samples_per_second": 23.972, | |
| "eval_steps_per_second": 3.001, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.3297872340425532, | |
| "grad_norm": 2.336822509765625, | |
| "learning_rate": 0.00029699999999999996, | |
| "loss": 9.6222, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.3297872340425532, | |
| "eval_loss": 4.171350479125977, | |
| "eval_runtime": 206.1998, | |
| "eval_samples_per_second": 24.059, | |
| "eval_steps_per_second": 3.012, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.5957446808510638, | |
| "eval_loss": 4.168086051940918, | |
| "eval_runtime": 206.1519, | |
| "eval_samples_per_second": 24.065, | |
| "eval_steps_per_second": 3.012, | |
| "eval_wer": 1.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.8617021276595744, | |
| "eval_loss": 4.133743762969971, | |
| "eval_runtime": 206.5072, | |
| "eval_samples_per_second": 24.023, | |
| "eval_steps_per_second": 3.007, | |
| "eval_wer": 1.0, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.127659574468085, | |
| "eval_loss": 4.110419273376465, | |
| "eval_runtime": 206.2896, | |
| "eval_samples_per_second": 24.049, | |
| "eval_steps_per_second": 3.01, | |
| "eval_wer": 1.0, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.393617021276596, | |
| "eval_loss": 4.1108503341674805, | |
| "eval_runtime": 206.9777, | |
| "eval_samples_per_second": 23.969, | |
| "eval_steps_per_second": 3.0, | |
| "eval_wer": 1.0, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.6595744680851063, | |
| "grad_norm": 1.0139914751052856, | |
| "learning_rate": 0.0002711089494163424, | |
| "loss": 4.1243, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.6595744680851063, | |
| "eval_loss": 4.111221790313721, | |
| "eval_runtime": 207.2581, | |
| "eval_samples_per_second": 23.936, | |
| "eval_steps_per_second": 2.996, | |
| "eval_wer": 1.0, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.925531914893617, | |
| "eval_loss": 4.116605758666992, | |
| "eval_runtime": 206.7383, | |
| "eval_samples_per_second": 23.997, | |
| "eval_steps_per_second": 3.004, | |
| "eval_wer": 1.0, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.1914893617021276, | |
| "eval_loss": 4.143283367156982, | |
| "eval_runtime": 207.4325, | |
| "eval_samples_per_second": 23.916, | |
| "eval_steps_per_second": 2.994, | |
| "eval_wer": 1.0, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.4574468085106385, | |
| "eval_loss": 4.188790321350098, | |
| "eval_runtime": 206.5637, | |
| "eval_samples_per_second": 24.017, | |
| "eval_steps_per_second": 3.006, | |
| "eval_wer": 1.0, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.723404255319149, | |
| "eval_loss": 4.2255859375, | |
| "eval_runtime": 206.5284, | |
| "eval_samples_per_second": 24.021, | |
| "eval_steps_per_second": 3.007, | |
| "eval_wer": 1.0, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.9893617021276597, | |
| "grad_norm": 2.125878095626831, | |
| "learning_rate": 0.0002419260700389105, | |
| "loss": 4.1253, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.9893617021276597, | |
| "eval_loss": 4.14175271987915, | |
| "eval_runtime": 206.1478, | |
| "eval_samples_per_second": 24.065, | |
| "eval_steps_per_second": 3.012, | |
| "eval_wer": 1.0, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.25531914893617, | |
| "eval_loss": 4.113326549530029, | |
| "eval_runtime": 206.8791, | |
| "eval_samples_per_second": 23.98, | |
| "eval_steps_per_second": 3.002, | |
| "eval_wer": 1.0, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.5212765957446805, | |
| "eval_loss": 4.112410545349121, | |
| "eval_runtime": 207.4656, | |
| "eval_samples_per_second": 23.912, | |
| "eval_steps_per_second": 2.993, | |
| "eval_wer": 1.0, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 4.787234042553192, | |
| "eval_loss": 4.113221645355225, | |
| "eval_runtime": 206.6651, | |
| "eval_samples_per_second": 24.005, | |
| "eval_steps_per_second": 3.005, | |
| "eval_wer": 1.0, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 5.053191489361702, | |
| "eval_loss": 4.19992208480835, | |
| "eval_runtime": 206.9983, | |
| "eval_samples_per_second": 23.966, | |
| "eval_steps_per_second": 3.0, | |
| "eval_wer": 1.0, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 5.319148936170213, | |
| "grad_norm": 1.172359585762024, | |
| "learning_rate": 0.0002127431906614786, | |
| "loss": 4.1182, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.319148936170213, | |
| "eval_loss": 4.179160118103027, | |
| "eval_runtime": 207.0024, | |
| "eval_samples_per_second": 23.966, | |
| "eval_steps_per_second": 3.0, | |
| "eval_wer": 1.0, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.585106382978723, | |
| "eval_loss": 4.211320400238037, | |
| "eval_runtime": 208.18, | |
| "eval_samples_per_second": 23.83, | |
| "eval_steps_per_second": 2.983, | |
| "eval_wer": 1.0, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 5.851063829787234, | |
| "eval_loss": 4.20604133605957, | |
| "eval_runtime": 207.2193, | |
| "eval_samples_per_second": 23.941, | |
| "eval_steps_per_second": 2.997, | |
| "eval_wer": 1.0, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 6.117021276595745, | |
| "eval_loss": 4.110590934753418, | |
| "eval_runtime": 208.0433, | |
| "eval_samples_per_second": 23.846, | |
| "eval_steps_per_second": 2.985, | |
| "eval_wer": 1.0, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 6.382978723404255, | |
| "eval_loss": 4.112478256225586, | |
| "eval_runtime": 207.5552, | |
| "eval_samples_per_second": 23.902, | |
| "eval_steps_per_second": 2.992, | |
| "eval_wer": 1.0, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 6.648936170212766, | |
| "grad_norm": 1.6996749639511108, | |
| "learning_rate": 0.0001835603112840467, | |
| "loss": 4.1207, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.648936170212766, | |
| "eval_loss": 4.113041877746582, | |
| "eval_runtime": 207.168, | |
| "eval_samples_per_second": 23.947, | |
| "eval_steps_per_second": 2.998, | |
| "eval_wer": 1.0, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.914893617021277, | |
| "eval_loss": 4.108416557312012, | |
| "eval_runtime": 207.1242, | |
| "eval_samples_per_second": 23.952, | |
| "eval_steps_per_second": 2.998, | |
| "eval_wer": 1.0, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 7.180851063829787, | |
| "eval_loss": 4.170561790466309, | |
| "eval_runtime": 210.6386, | |
| "eval_samples_per_second": 23.552, | |
| "eval_steps_per_second": 2.948, | |
| "eval_wer": 1.0, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 7.446808510638298, | |
| "eval_loss": 4.179252624511719, | |
| "eval_runtime": 207.7058, | |
| "eval_samples_per_second": 23.885, | |
| "eval_steps_per_second": 2.99, | |
| "eval_wer": 1.0, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 7.712765957446808, | |
| "eval_loss": 4.18436336517334, | |
| "eval_runtime": 207.7366, | |
| "eval_samples_per_second": 23.881, | |
| "eval_steps_per_second": 2.989, | |
| "eval_wer": 1.0, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 7.9787234042553195, | |
| "grad_norm": 3.874195098876953, | |
| "learning_rate": 0.0001544941634241245, | |
| "loss": 4.3809, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 7.9787234042553195, | |
| "eval_loss": 4.143243789672852, | |
| "eval_runtime": 208.1415, | |
| "eval_samples_per_second": 23.835, | |
| "eval_steps_per_second": 2.984, | |
| "eval_wer": 1.0, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.24468085106383, | |
| "eval_loss": 4.108314514160156, | |
| "eval_runtime": 208.2271, | |
| "eval_samples_per_second": 23.825, | |
| "eval_steps_per_second": 2.982, | |
| "eval_wer": 1.0, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 8.51063829787234, | |
| "eval_loss": 4.117116928100586, | |
| "eval_runtime": 208.0587, | |
| "eval_samples_per_second": 23.844, | |
| "eval_steps_per_second": 2.985, | |
| "eval_wer": 1.0, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 8.77659574468085, | |
| "eval_loss": 4.11330509185791, | |
| "eval_runtime": 208.1596, | |
| "eval_samples_per_second": 23.833, | |
| "eval_steps_per_second": 2.983, | |
| "eval_wer": 1.0, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 9.042553191489361, | |
| "eval_loss": 4.163766384124756, | |
| "eval_runtime": 208.0242, | |
| "eval_samples_per_second": 23.848, | |
| "eval_steps_per_second": 2.985, | |
| "eval_wer": 1.0, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 9.308510638297872, | |
| "grad_norm": 2.0024287700653076, | |
| "learning_rate": 0.0001253112840466926, | |
| "loss": 4.1164, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.308510638297872, | |
| "eval_loss": 4.130943298339844, | |
| "eval_runtime": 208.5062, | |
| "eval_samples_per_second": 23.793, | |
| "eval_steps_per_second": 2.978, | |
| "eval_wer": 1.0, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.574468085106384, | |
| "eval_loss": 4.1129841804504395, | |
| "eval_runtime": 207.7504, | |
| "eval_samples_per_second": 23.88, | |
| "eval_steps_per_second": 2.989, | |
| "eval_wer": 1.0, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 9.840425531914894, | |
| "eval_loss": 4.127650737762451, | |
| "eval_runtime": 208.5195, | |
| "eval_samples_per_second": 23.792, | |
| "eval_steps_per_second": 2.978, | |
| "eval_wer": 1.0, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 10.106382978723405, | |
| "eval_loss": 4.1156158447265625, | |
| "eval_runtime": 208.3668, | |
| "eval_samples_per_second": 23.809, | |
| "eval_steps_per_second": 2.98, | |
| "eval_wer": 1.0, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 10.372340425531915, | |
| "eval_loss": 4.1260600090026855, | |
| "eval_runtime": 208.8761, | |
| "eval_samples_per_second": 23.751, | |
| "eval_steps_per_second": 2.973, | |
| "eval_wer": 1.0, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 10.638297872340425, | |
| "grad_norm": 1.807287573814392, | |
| "learning_rate": 9.61284046692607e-05, | |
| "loss": 4.1195, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.638297872340425, | |
| "eval_loss": 4.130354404449463, | |
| "eval_runtime": 208.2427, | |
| "eval_samples_per_second": 23.823, | |
| "eval_steps_per_second": 2.982, | |
| "eval_wer": 1.0, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.904255319148936, | |
| "eval_loss": 4.128208637237549, | |
| "eval_runtime": 208.9946, | |
| "eval_samples_per_second": 23.737, | |
| "eval_steps_per_second": 2.971, | |
| "eval_wer": 1.0, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 11.170212765957446, | |
| "eval_loss": 4.11016845703125, | |
| "eval_runtime": 214.217, | |
| "eval_samples_per_second": 23.159, | |
| "eval_steps_per_second": 2.899, | |
| "eval_wer": 1.0, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 11.436170212765958, | |
| "eval_loss": 4.111141681671143, | |
| "eval_runtime": 208.356, | |
| "eval_samples_per_second": 23.81, | |
| "eval_steps_per_second": 2.98, | |
| "eval_wer": 1.0, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 11.702127659574469, | |
| "eval_loss": 4.111063003540039, | |
| "eval_runtime": 209.2862, | |
| "eval_samples_per_second": 23.704, | |
| "eval_steps_per_second": 2.967, | |
| "eval_wer": 1.0, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 11.96808510638298, | |
| "grad_norm": 2.2066569328308105, | |
| "learning_rate": 6.69455252918288e-05, | |
| "loss": 4.1164, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 11.96808510638298, | |
| "eval_loss": 4.112732410430908, | |
| "eval_runtime": 213.9347, | |
| "eval_samples_per_second": 23.189, | |
| "eval_steps_per_second": 2.903, | |
| "eval_wer": 1.0, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 12.23404255319149, | |
| "eval_loss": 4.141692638397217, | |
| "eval_runtime": 214.7394, | |
| "eval_samples_per_second": 23.102, | |
| "eval_steps_per_second": 2.892, | |
| "eval_wer": 1.0, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "eval_loss": 4.139012336730957, | |
| "eval_runtime": 223.6932, | |
| "eval_samples_per_second": 22.178, | |
| "eval_steps_per_second": 2.776, | |
| "eval_wer": 1.0, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 12.76595744680851, | |
| "eval_loss": 4.143847942352295, | |
| "eval_runtime": 207.9672, | |
| "eval_samples_per_second": 23.855, | |
| "eval_steps_per_second": 2.986, | |
| "eval_wer": 1.0, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 13.03191489361702, | |
| "eval_loss": 4.122192859649658, | |
| "eval_runtime": 208.5333, | |
| "eval_samples_per_second": 23.79, | |
| "eval_steps_per_second": 2.978, | |
| "eval_wer": 1.0, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 13.297872340425531, | |
| "grad_norm": 0.3912961781024933, | |
| "learning_rate": 3.776264591439688e-05, | |
| "loss": 4.1117, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 13.297872340425531, | |
| "eval_loss": 4.115208625793457, | |
| "eval_runtime": 208.8419, | |
| "eval_samples_per_second": 23.755, | |
| "eval_steps_per_second": 2.974, | |
| "eval_wer": 1.0, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 13.563829787234042, | |
| "eval_loss": 4.124714374542236, | |
| "eval_runtime": 208.9077, | |
| "eval_samples_per_second": 23.747, | |
| "eval_steps_per_second": 2.973, | |
| "eval_wer": 1.0, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 13.829787234042554, | |
| "eval_loss": 4.124392032623291, | |
| "eval_runtime": 208.9383, | |
| "eval_samples_per_second": 23.744, | |
| "eval_steps_per_second": 2.972, | |
| "eval_wer": 1.0, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 14.095744680851064, | |
| "eval_loss": 4.138058662414551, | |
| "eval_runtime": 208.8904, | |
| "eval_samples_per_second": 23.749, | |
| "eval_steps_per_second": 2.973, | |
| "eval_wer": 1.0, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 14.361702127659575, | |
| "eval_loss": 4.13184928894043, | |
| "eval_runtime": 209.1322, | |
| "eval_samples_per_second": 23.722, | |
| "eval_steps_per_second": 2.969, | |
| "eval_wer": 1.0, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 14.627659574468085, | |
| "grad_norm": 1.6473333835601807, | |
| "learning_rate": 8.579766536964979e-06, | |
| "loss": 4.1079, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 14.627659574468085, | |
| "eval_loss": 4.131294250488281, | |
| "eval_runtime": 209.206, | |
| "eval_samples_per_second": 23.713, | |
| "eval_steps_per_second": 2.968, | |
| "eval_wer": 1.0, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 14.893617021276595, | |
| "eval_loss": 4.130970478057861, | |
| "eval_runtime": 209.3258, | |
| "eval_samples_per_second": 23.7, | |
| "eval_steps_per_second": 2.967, | |
| "eval_wer": 1.0, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 5640, | |
| "total_flos": 7.889727743335047e+18, | |
| "train_loss": 4.629210873867603, | |
| "train_runtime": 29472.0498, | |
| "train_samples_per_second": 6.124, | |
| "train_steps_per_second": 0.191 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 5640, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "save_steps": 400, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.889727743335047e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |