| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.997489539748955, | |
| "global_step": 2980, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.9000000000000005e-05, | |
| "loss": 5.3579, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 9.900000000000001e-05, | |
| "loss": 3.2697, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 0.000149, | |
| "loss": 2.4711, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.000199, | |
| "loss": 2.1202, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.000249, | |
| "loss": 2.0846, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_loss": 1.1640619039535522, | |
| "eval_runtime": 414.2546, | |
| "eval_samples_per_second": 25.076, | |
| "eval_steps_per_second": 3.136, | |
| "eval_wer": 0.8072035074088462, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 0.000299, | |
| "loss": 2.1202, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 0.00034899999999999997, | |
| "loss": 2.1037, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 0.00039900000000000005, | |
| "loss": 2.0882, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 0.000449, | |
| "loss": 2.126, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 0.000499, | |
| "loss": 2.1201, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "eval_loss": 1.1776171922683716, | |
| "eval_runtime": 424.2371, | |
| "eval_samples_per_second": 24.486, | |
| "eval_steps_per_second": 3.062, | |
| "eval_wer": 0.8329171060177221, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 0.000549, | |
| "loss": 2.1435, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 0.000599, | |
| "loss": 2.152, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 0.0006490000000000001, | |
| "loss": 2.11, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 0.000699, | |
| "loss": 2.1503, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 0.000749, | |
| "loss": 2.1972, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "eval_loss": 1.2631869316101074, | |
| "eval_runtime": 410.677, | |
| "eval_samples_per_second": 25.295, | |
| "eval_steps_per_second": 3.163, | |
| "eval_wer": 0.8723754555376732, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.37, | |
| "learning_rate": 0.000799, | |
| "loss": 2.1788, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "learning_rate": 0.000849, | |
| "loss": 2.1774, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 0.0008990000000000001, | |
| "loss": 2.2205, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 6.37, | |
| "learning_rate": 0.000949, | |
| "loss": 2.2466, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 6.71, | |
| "learning_rate": 0.000999, | |
| "loss": 2.2643, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.71, | |
| "eval_loss": 1.372324824333191, | |
| "eval_runtime": 411.6031, | |
| "eval_samples_per_second": 25.238, | |
| "eval_steps_per_second": 3.156, | |
| "eval_wer": 0.8982740440645984, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 0.0009000000000000001, | |
| "loss": 2.2996, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 7.38, | |
| "learning_rate": 0.0007979591836734694, | |
| "loss": 2.2775, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "learning_rate": 0.0006959183673469388, | |
| "loss": 2.2478, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "learning_rate": 0.0005938775510204082, | |
| "loss": 2.2288, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 8.39, | |
| "learning_rate": 0.0004918367346938776, | |
| "loss": 2.1649, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.39, | |
| "eval_loss": 1.25503671169281, | |
| "eval_runtime": 416.24, | |
| "eval_samples_per_second": 24.957, | |
| "eval_steps_per_second": 3.121, | |
| "eval_wer": 0.884196311301034, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.72, | |
| "learning_rate": 0.000389795918367347, | |
| "loss": 2.1054, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "learning_rate": 0.00028775510204081633, | |
| "loss": 2.0872, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 9.39, | |
| "learning_rate": 0.00018571428571428572, | |
| "loss": 1.9953, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 9.73, | |
| "learning_rate": 8.36734693877551e-05, | |
| "loss": 1.9474, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 2980, | |
| "total_flos": 1.006981652948512e+20, | |
| "train_loss": 2.3014913469352978, | |
| "train_runtime": 21335.6515, | |
| "train_samples_per_second": 17.909, | |
| "train_steps_per_second": 0.14 | |
| } | |
| ], | |
| "max_steps": 2980, | |
| "num_train_epochs": 10, | |
| "total_flos": 1.006981652948512e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |