| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.5209125475285172, | |
| "eval_steps": 100, | |
| "global_step": 1600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.2933333333333334e-06, | |
| "loss": 19.0829, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 19.174156188964844, | |
| "eval_runtime": 65.9422, | |
| "eval_samples_per_second": 6.491, | |
| "eval_steps_per_second": 0.819, | |
| "eval_wer": 1.0007501875468867, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 2.6e-06, | |
| "loss": 18.8941, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_loss": 18.407485961914062, | |
| "eval_runtime": 66.0283, | |
| "eval_samples_per_second": 6.482, | |
| "eval_steps_per_second": 0.818, | |
| "eval_wer": 1.009627406851713, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 3.9333333333333335e-06, | |
| "loss": 17.5359, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 17.08553123474121, | |
| "eval_runtime": 65.7138, | |
| "eval_samples_per_second": 6.513, | |
| "eval_steps_per_second": 0.822, | |
| "eval_wer": 1.248312078019505, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 5.2666666666666665e-06, | |
| "loss": 16.0015, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "eval_loss": 15.136056900024414, | |
| "eval_runtime": 65.1801, | |
| "eval_samples_per_second": 6.566, | |
| "eval_steps_per_second": 0.828, | |
| "eval_wer": 1.3134533633408352, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 6.5866666666666666e-06, | |
| "loss": 13.3605, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_loss": 12.500285148620605, | |
| "eval_runtime": 65.5504, | |
| "eval_samples_per_second": 6.529, | |
| "eval_steps_per_second": 0.824, | |
| "eval_wer": 1.0840210052513128, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 7.92e-06, | |
| "loss": 10.8925, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_loss": 9.41340160369873, | |
| "eval_runtime": 64.6713, | |
| "eval_samples_per_second": 6.618, | |
| "eval_steps_per_second": 0.835, | |
| "eval_wer": 1.0077519379844961, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 9.253333333333333e-06, | |
| "loss": 8.1129, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_loss": 6.854950904846191, | |
| "eval_runtime": 65.9954, | |
| "eval_samples_per_second": 6.485, | |
| "eval_steps_per_second": 0.818, | |
| "eval_wer": 1.0, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 1.0586666666666668e-05, | |
| "loss": 6.0586, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_loss": 5.390443801879883, | |
| "eval_runtime": 65.312, | |
| "eval_samples_per_second": 6.553, | |
| "eval_steps_per_second": 0.827, | |
| "eval_wer": 1.0, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 1.1920000000000001e-05, | |
| "loss": 5.1935, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "eval_loss": 4.793299674987793, | |
| "eval_runtime": 66.778, | |
| "eval_samples_per_second": 6.409, | |
| "eval_steps_per_second": 0.809, | |
| "eval_wer": 1.0, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 1.3253333333333334e-05, | |
| "loss": 4.7315, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 4.52764368057251, | |
| "eval_runtime": 66.1481, | |
| "eval_samples_per_second": 6.47, | |
| "eval_steps_per_second": 0.816, | |
| "eval_wer": 0.9998749687421855, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 1.4586666666666667e-05, | |
| "loss": 4.4273, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "eval_loss": 4.082287788391113, | |
| "eval_runtime": 65.6019, | |
| "eval_samples_per_second": 6.524, | |
| "eval_steps_per_second": 0.823, | |
| "eval_wer": 1.0303825956489123, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 1.5920000000000003e-05, | |
| "loss": 4.0406, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "eval_loss": 3.9247243404388428, | |
| "eval_runtime": 64.106, | |
| "eval_samples_per_second": 6.676, | |
| "eval_steps_per_second": 0.842, | |
| "eval_wer": 1.0102525631407853, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 1.7253333333333336e-05, | |
| "loss": 3.9165, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_loss": 3.870806932449341, | |
| "eval_runtime": 65.1335, | |
| "eval_samples_per_second": 6.571, | |
| "eval_steps_per_second": 0.829, | |
| "eval_wer": 1.0046261565391348, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 1.858666666666667e-05, | |
| "loss": 3.9965, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "eval_loss": 3.8457767963409424, | |
| "eval_runtime": 65.82, | |
| "eval_samples_per_second": 6.503, | |
| "eval_steps_per_second": 0.82, | |
| "eval_wer": 1.0012503125781445, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 1.9920000000000002e-05, | |
| "loss": 3.8683, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "eval_loss": 3.8101820945739746, | |
| "eval_runtime": 65.6088, | |
| "eval_samples_per_second": 6.524, | |
| "eval_steps_per_second": 0.823, | |
| "eval_wer": 1.0, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 2.1253333333333335e-05, | |
| "loss": 3.8863, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "eval_loss": 3.795001268386841, | |
| "eval_runtime": 65.0097, | |
| "eval_samples_per_second": 6.584, | |
| "eval_steps_per_second": 0.831, | |
| "eval_wer": 1.0037509377344336, | |
| "step": 1600 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 21040, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 200, | |
| "total_flos": 3.1104086860681257e+18, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |