| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 50.0, | |
| "global_step": 2550, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 7.361999999999999e-05, | |
| "loss": 10.1224, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "eval_loss": 3.5429043769836426, | |
| "eval_runtime": 12.8681, | |
| "eval_samples_per_second": 26.5, | |
| "eval_steps_per_second": 3.342, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 7.062e-05, | |
| "loss": 3.2411, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "eval_loss": 3.1785788536071777, | |
| "eval_runtime": 12.8876, | |
| "eval_samples_per_second": 26.46, | |
| "eval_steps_per_second": 3.337, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "learning_rate": 6.761999999999999e-05, | |
| "loss": 3.1283, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "eval_loss": 3.057115316390991, | |
| "eval_runtime": 12.9515, | |
| "eval_samples_per_second": 26.329, | |
| "eval_steps_per_second": 3.32, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.84, | |
| "learning_rate": 6.462e-05, | |
| "loss": 3.0044, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 7.84, | |
| "eval_loss": 2.956029176712036, | |
| "eval_runtime": 13.0792, | |
| "eval_samples_per_second": 26.072, | |
| "eval_steps_per_second": 3.288, | |
| "eval_wer": 0.9996309963099631, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 9.8, | |
| "learning_rate": 6.162e-05, | |
| "loss": 2.9388, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 9.8, | |
| "eval_loss": 2.897726535797119, | |
| "eval_runtime": 13.249, | |
| "eval_samples_per_second": 25.738, | |
| "eval_steps_per_second": 3.246, | |
| "eval_wer": 1.0011070110701108, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 11.76, | |
| "learning_rate": 5.861999999999999e-05, | |
| "loss": 2.86, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 11.76, | |
| "eval_loss": 2.694431781768799, | |
| "eval_runtime": 12.8219, | |
| "eval_samples_per_second": 26.595, | |
| "eval_steps_per_second": 3.354, | |
| "eval_wer": 0.9952029520295202, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 13.73, | |
| "learning_rate": 5.562e-05, | |
| "loss": 2.5538, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 13.73, | |
| "eval_loss": 2.0967445373535156, | |
| "eval_runtime": 12.8896, | |
| "eval_samples_per_second": 26.456, | |
| "eval_steps_per_second": 3.336, | |
| "eval_wer": 0.9435424354243542, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 15.69, | |
| "learning_rate": 5.2619999999999994e-05, | |
| "loss": 2.1214, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 15.69, | |
| "eval_loss": 1.4815592765808105, | |
| "eval_runtime": 12.9225, | |
| "eval_samples_per_second": 26.388, | |
| "eval_steps_per_second": 3.328, | |
| "eval_wer": 0.8428044280442805, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 17.65, | |
| "learning_rate": 4.9619999999999996e-05, | |
| "loss": 1.8136, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 17.65, | |
| "eval_loss": 1.245869517326355, | |
| "eval_runtime": 12.7527, | |
| "eval_samples_per_second": 26.739, | |
| "eval_steps_per_second": 3.372, | |
| "eval_wer": 0.8047970479704797, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 19.61, | |
| "learning_rate": 4.662e-05, | |
| "loss": 1.6795, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 19.61, | |
| "eval_loss": 1.1232149600982666, | |
| "eval_runtime": 12.9316, | |
| "eval_samples_per_second": 26.37, | |
| "eval_steps_per_second": 3.325, | |
| "eval_wer": 0.7649446494464944, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 21.57, | |
| "learning_rate": 4.362e-05, | |
| "loss": 1.5571, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 21.57, | |
| "eval_loss": 1.0510376691818237, | |
| "eval_runtime": 12.7516, | |
| "eval_samples_per_second": 26.742, | |
| "eval_steps_per_second": 3.372, | |
| "eval_wer": 0.7431734317343174, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 23.53, | |
| "learning_rate": 4.0619999999999994e-05, | |
| "loss": 1.4975, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 23.53, | |
| "eval_loss": 1.0297844409942627, | |
| "eval_runtime": 12.7119, | |
| "eval_samples_per_second": 26.825, | |
| "eval_steps_per_second": 3.383, | |
| "eval_wer": 0.696309963099631, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 25.49, | |
| "learning_rate": 3.762e-05, | |
| "loss": 1.4485, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 25.49, | |
| "eval_loss": 0.9775241017341614, | |
| "eval_runtime": 12.651, | |
| "eval_samples_per_second": 26.954, | |
| "eval_steps_per_second": 3.399, | |
| "eval_wer": 0.707380073800738, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 27.45, | |
| "learning_rate": 3.462e-05, | |
| "loss": 1.3924, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 27.45, | |
| "eval_loss": 0.9797706007957458, | |
| "eval_runtime": 12.6253, | |
| "eval_samples_per_second": 27.009, | |
| "eval_steps_per_second": 3.406, | |
| "eval_wer": 0.6955719557195572, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 29.41, | |
| "learning_rate": 3.161999999999999e-05, | |
| "loss": 1.3604, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 29.41, | |
| "eval_loss": 0.9344653487205505, | |
| "eval_runtime": 12.7334, | |
| "eval_samples_per_second": 26.78, | |
| "eval_steps_per_second": 3.377, | |
| "eval_wer": 0.7092250922509226, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 31.37, | |
| "learning_rate": 2.8619999999999997e-05, | |
| "loss": 1.3224, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 31.37, | |
| "eval_loss": 0.9535229802131653, | |
| "eval_runtime": 12.7291, | |
| "eval_samples_per_second": 26.789, | |
| "eval_steps_per_second": 3.378, | |
| "eval_wer": 0.6830258302583025, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "learning_rate": 2.562e-05, | |
| "loss": 1.2816, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "eval_loss": 0.9177776575088501, | |
| "eval_runtime": 12.8107, | |
| "eval_samples_per_second": 26.618, | |
| "eval_steps_per_second": 3.357, | |
| "eval_wer": 0.6678966789667896, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 35.29, | |
| "learning_rate": 2.2619999999999997e-05, | |
| "loss": 1.2623, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 35.29, | |
| "eval_loss": 0.924865186214447, | |
| "eval_runtime": 12.5463, | |
| "eval_samples_per_second": 27.179, | |
| "eval_steps_per_second": 3.427, | |
| "eval_wer": 0.6678966789667896, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 37.25, | |
| "learning_rate": 1.962e-05, | |
| "loss": 1.2421, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 37.25, | |
| "eval_loss": 0.9123861193656921, | |
| "eval_runtime": 12.7025, | |
| "eval_samples_per_second": 26.845, | |
| "eval_steps_per_second": 3.385, | |
| "eval_wer": 0.6734317343173432, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 39.22, | |
| "learning_rate": 1.6619999999999997e-05, | |
| "loss": 1.2208, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 39.22, | |
| "eval_loss": 0.8961586356163025, | |
| "eval_runtime": 12.5233, | |
| "eval_samples_per_second": 27.229, | |
| "eval_steps_per_second": 3.434, | |
| "eval_wer": 0.666420664206642, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 41.18, | |
| "learning_rate": 1.362e-05, | |
| "loss": 1.2145, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 41.18, | |
| "eval_loss": 0.8903014063835144, | |
| "eval_runtime": 12.6636, | |
| "eval_samples_per_second": 26.928, | |
| "eval_steps_per_second": 3.396, | |
| "eval_wer": 0.6734317343173432, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 43.14, | |
| "learning_rate": 1.062e-05, | |
| "loss": 1.1888, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 43.14, | |
| "eval_loss": 0.8883377313613892, | |
| "eval_runtime": 12.7667, | |
| "eval_samples_per_second": 26.71, | |
| "eval_steps_per_second": 3.368, | |
| "eval_wer": 0.6708487084870849, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 45.1, | |
| "learning_rate": 7.619999999999999e-06, | |
| "loss": 1.1933, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 45.1, | |
| "eval_loss": 0.8928494453430176, | |
| "eval_runtime": 12.6877, | |
| "eval_samples_per_second": 26.876, | |
| "eval_steps_per_second": 3.389, | |
| "eval_wer": 0.6723247232472325, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 47.06, | |
| "learning_rate": 4.62e-06, | |
| "loss": 1.1838, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 47.06, | |
| "eval_loss": 0.8868067860603333, | |
| "eval_runtime": 12.5901, | |
| "eval_samples_per_second": 27.085, | |
| "eval_steps_per_second": 3.415, | |
| "eval_wer": 0.6678966789667896, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 49.02, | |
| "learning_rate": 1.62e-06, | |
| "loss": 1.1634, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 49.02, | |
| "eval_loss": 0.888584554195404, | |
| "eval_runtime": 12.7157, | |
| "eval_samples_per_second": 26.817, | |
| "eval_steps_per_second": 3.382, | |
| "eval_wer": 0.6656826568265682, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 2550, | |
| "total_flos": 4.920553620909019e+18, | |
| "train_loss": 2.1006999116785385, | |
| "train_runtime": 2771.7539, | |
| "train_samples_per_second": 14.612, | |
| "train_steps_per_second": 0.92 | |
| } | |
| ], | |
| "max_steps": 2550, | |
| "num_train_epochs": 50, | |
| "total_flos": 4.920553620909019e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |