| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.536067892503536, | |
| "eval_steps": 100, | |
| "global_step": 2500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14144271570014144, | |
| "eval_loss": 6.970242977142334, | |
| "eval_runtime": 137.617, | |
| "eval_samples_per_second": 41.1, | |
| "eval_steps_per_second": 5.137, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2828854314002829, | |
| "eval_loss": 3.36755633354187, | |
| "eval_runtime": 133.5904, | |
| "eval_samples_per_second": 42.338, | |
| "eval_steps_per_second": 5.292, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4243281471004243, | |
| "eval_loss": 3.0392866134643555, | |
| "eval_runtime": 133.8525, | |
| "eval_samples_per_second": 42.255, | |
| "eval_steps_per_second": 5.282, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5657708628005658, | |
| "eval_loss": 2.9887070655822754, | |
| "eval_runtime": 134.092, | |
| "eval_samples_per_second": 42.18, | |
| "eval_steps_per_second": 5.272, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7072135785007072, | |
| "grad_norm": 0.6687237620353699, | |
| "learning_rate": 0.00029699999999999996, | |
| "loss": 4.717, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.7072135785007072, | |
| "eval_loss": 3.0226564407348633, | |
| "eval_runtime": 134.7626, | |
| "eval_samples_per_second": 41.97, | |
| "eval_steps_per_second": 5.246, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8486562942008486, | |
| "eval_loss": 3.0405795574188232, | |
| "eval_runtime": 135.9502, | |
| "eval_samples_per_second": 41.603, | |
| "eval_steps_per_second": 5.2, | |
| "eval_wer": 1.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9900990099009901, | |
| "eval_loss": 3.0028798580169678, | |
| "eval_runtime": 135.0397, | |
| "eval_samples_per_second": 41.884, | |
| "eval_steps_per_second": 5.235, | |
| "eval_wer": 1.0, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.1315417256011315, | |
| "eval_loss": 2.948317050933838, | |
| "eval_runtime": 135.3684, | |
| "eval_samples_per_second": 41.782, | |
| "eval_steps_per_second": 5.223, | |
| "eval_wer": 1.0, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.272984441301273, | |
| "eval_loss": 2.9510738849639893, | |
| "eval_runtime": 135.3897, | |
| "eval_samples_per_second": 41.776, | |
| "eval_steps_per_second": 5.222, | |
| "eval_wer": 1.0, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.4144271570014144, | |
| "grad_norm": 0.3827725648880005, | |
| "learning_rate": 0.00022574999999999996, | |
| "loss": 3.0065, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.4144271570014144, | |
| "eval_loss": 2.9472572803497314, | |
| "eval_runtime": 134.8251, | |
| "eval_samples_per_second": 41.951, | |
| "eval_steps_per_second": 5.244, | |
| "eval_wer": 1.0, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.5558698727015559, | |
| "eval_loss": 2.9448139667510986, | |
| "eval_runtime": 135.1998, | |
| "eval_samples_per_second": 41.834, | |
| "eval_steps_per_second": 5.229, | |
| "eval_wer": 1.0, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.6973125884016973, | |
| "eval_loss": 2.946981191635132, | |
| "eval_runtime": 135.1335, | |
| "eval_samples_per_second": 41.855, | |
| "eval_steps_per_second": 5.232, | |
| "eval_wer": 1.0, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.8387553041018387, | |
| "eval_loss": 2.944624662399292, | |
| "eval_runtime": 135.5596, | |
| "eval_samples_per_second": 41.723, | |
| "eval_steps_per_second": 5.215, | |
| "eval_wer": 1.0, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.9801980198019802, | |
| "eval_loss": 2.9431614875793457, | |
| "eval_runtime": 135.6031, | |
| "eval_samples_per_second": 41.71, | |
| "eval_steps_per_second": 5.214, | |
| "eval_wer": 1.0, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.1216407355021216, | |
| "grad_norm": 0.5749518275260925, | |
| "learning_rate": 0.0001512, | |
| "loss": 2.9634, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.1216407355021216, | |
| "eval_loss": 2.9475975036621094, | |
| "eval_runtime": 136.0526, | |
| "eval_samples_per_second": 41.572, | |
| "eval_steps_per_second": 5.197, | |
| "eval_wer": 1.0, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.263083451202263, | |
| "eval_loss": 2.9624321460723877, | |
| "eval_runtime": 136.0863, | |
| "eval_samples_per_second": 41.562, | |
| "eval_steps_per_second": 5.195, | |
| "eval_wer": 1.0, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.4045261669024045, | |
| "eval_loss": 2.9580626487731934, | |
| "eval_runtime": 136.566, | |
| "eval_samples_per_second": 41.416, | |
| "eval_steps_per_second": 5.177, | |
| "eval_wer": 1.0, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.545968882602546, | |
| "eval_loss": 2.9552838802337646, | |
| "eval_runtime": 136.3048, | |
| "eval_samples_per_second": 41.495, | |
| "eval_steps_per_second": 5.187, | |
| "eval_wer": 1.0, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.6874115983026874, | |
| "eval_loss": 2.9515163898468018, | |
| "eval_runtime": 136.0496, | |
| "eval_samples_per_second": 41.573, | |
| "eval_steps_per_second": 5.197, | |
| "eval_wer": 1.0, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.828854314002829, | |
| "grad_norm": 0.206673726439476, | |
| "learning_rate": 7.664999999999999e-05, | |
| "loss": 2.9677, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.828854314002829, | |
| "eval_loss": 2.9480981826782227, | |
| "eval_runtime": 134.8264, | |
| "eval_samples_per_second": 41.95, | |
| "eval_steps_per_second": 5.244, | |
| "eval_wer": 1.0, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.9702970297029703, | |
| "eval_loss": 2.9509432315826416, | |
| "eval_runtime": 134.773, | |
| "eval_samples_per_second": 41.967, | |
| "eval_steps_per_second": 5.246, | |
| "eval_wer": 1.0, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.1117397454031117, | |
| "eval_loss": 2.940824031829834, | |
| "eval_runtime": 135.008, | |
| "eval_samples_per_second": 41.894, | |
| "eval_steps_per_second": 5.237, | |
| "eval_wer": 1.0, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.253182461103253, | |
| "eval_loss": 2.9393398761749268, | |
| "eval_runtime": 134.9605, | |
| "eval_samples_per_second": 41.909, | |
| "eval_steps_per_second": 5.239, | |
| "eval_wer": 1.0, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 3.3946251768033946, | |
| "eval_loss": 2.938136339187622, | |
| "eval_runtime": 134.6374, | |
| "eval_samples_per_second": 42.009, | |
| "eval_steps_per_second": 5.251, | |
| "eval_wer": 1.0, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.536067892503536, | |
| "grad_norm": 0.6203744411468506, | |
| "learning_rate": 1.9499999999999995e-06, | |
| "loss": 2.9612, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.536067892503536, | |
| "eval_loss": 2.9374959468841553, | |
| "eval_runtime": 134.9253, | |
| "eval_samples_per_second": 41.919, | |
| "eval_steps_per_second": 5.24, | |
| "eval_wer": 1.0, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.536067892503536, | |
| "step": 2500, | |
| "total_flos": 8.753895488690287e+18, | |
| "train_loss": 3.32313603515625, | |
| "train_runtime": 6689.3091, | |
| "train_samples_per_second": 11.959, | |
| "train_steps_per_second": 0.374 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 400, | |
| "total_flos": 8.753895488690287e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |