| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 27.11864406779661, | |
| "global_step": 8000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.36, | |
| "eval_loss": 1.4595407247543335, | |
| "eval_runtime": 81.6295, | |
| "eval_samples_per_second": 25.665, | |
| "eval_steps_per_second": 3.21, | |
| "eval_wer": 1.0039260592180599, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.0002982, | |
| "loss": 4.7778, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "eval_loss": 0.8082281947135925, | |
| "eval_runtime": 80.8384, | |
| "eval_samples_per_second": 25.916, | |
| "eval_steps_per_second": 3.241, | |
| "eval_wer": 1.0115055346529254, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 0.00028011999999999997, | |
| "loss": 0.6408, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "eval_loss": 0.7031667828559875, | |
| "eval_runtime": 84.117, | |
| "eval_samples_per_second": 24.906, | |
| "eval_steps_per_second": 3.115, | |
| "eval_wer": 1.0078521184361198, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 0.00026011999999999997, | |
| "loss": 0.3937, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.42, | |
| "eval_loss": 0.6889204382896423, | |
| "eval_runtime": 80.1903, | |
| "eval_samples_per_second": 26.125, | |
| "eval_steps_per_second": 3.267, | |
| "eval_wer": 1.0432957085991603, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "learning_rate": 0.00024011999999999997, | |
| "loss": 0.3, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "eval_loss": 0.6820310354232788, | |
| "eval_runtime": 80.2232, | |
| "eval_samples_per_second": 26.115, | |
| "eval_steps_per_second": 3.266, | |
| "eval_wer": 1.0068706036316049, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 8.14, | |
| "eval_loss": 0.6669920086860657, | |
| "eval_runtime": 79.89, | |
| "eval_samples_per_second": 26.224, | |
| "eval_steps_per_second": 3.28, | |
| "eval_wer": 1.0196302960902994, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "learning_rate": 0.00022011999999999997, | |
| "loss": 0.226, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 9.49, | |
| "eval_loss": 0.7215595841407776, | |
| "eval_runtime": 80.0002, | |
| "eval_samples_per_second": 26.187, | |
| "eval_steps_per_second": 3.275, | |
| "eval_wer": 1.0422051365941436, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 10.17, | |
| "learning_rate": 0.00020012, | |
| "loss": 0.197, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 10.85, | |
| "eval_loss": 0.7669464945793152, | |
| "eval_runtime": 80.1105, | |
| "eval_samples_per_second": 26.151, | |
| "eval_steps_per_second": 3.27, | |
| "eval_wer": 1.053383499645564, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 11.86, | |
| "learning_rate": 0.00018012, | |
| "loss": 0.165, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 12.2, | |
| "eval_loss": 0.7517344951629639, | |
| "eval_runtime": 79.716, | |
| "eval_samples_per_second": 26.281, | |
| "eval_steps_per_second": 3.287, | |
| "eval_wer": 1.0199574676918044, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 13.56, | |
| "learning_rate": 0.00016011999999999998, | |
| "loss": 0.1486, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 13.56, | |
| "eval_loss": 0.7124771475791931, | |
| "eval_runtime": 79.8981, | |
| "eval_samples_per_second": 26.221, | |
| "eval_steps_per_second": 3.279, | |
| "eval_wer": 1.0357162331642946, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 14.92, | |
| "eval_loss": 0.7447456121444702, | |
| "eval_runtime": 82.2103, | |
| "eval_samples_per_second": 25.483, | |
| "eval_steps_per_second": 3.187, | |
| "eval_wer": 1.0347347183597797, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 15.25, | |
| "learning_rate": 0.00014012, | |
| "loss": 0.122, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 16.27, | |
| "eval_loss": 0.6899322271347046, | |
| "eval_runtime": 79.6624, | |
| "eval_samples_per_second": 26.298, | |
| "eval_steps_per_second": 3.289, | |
| "eval_wer": 1.0440045804024212, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 16.95, | |
| "learning_rate": 0.00012011999999999998, | |
| "loss": 0.1069, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 17.63, | |
| "eval_loss": 0.7212241291999817, | |
| "eval_runtime": 80.3141, | |
| "eval_samples_per_second": 26.085, | |
| "eval_steps_per_second": 3.262, | |
| "eval_wer": 1.0350073613610338, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 18.64, | |
| "learning_rate": 0.00010011999999999998, | |
| "loss": 0.0961, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 18.98, | |
| "eval_loss": 0.7417359352111816, | |
| "eval_runtime": 80.0211, | |
| "eval_samples_per_second": 26.181, | |
| "eval_steps_per_second": 3.274, | |
| "eval_wer": 1.0408419215878728, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 20.34, | |
| "learning_rate": 8.012e-05, | |
| "loss": 0.086, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 20.34, | |
| "eval_loss": 0.7402010560035706, | |
| "eval_runtime": 80.0522, | |
| "eval_samples_per_second": 26.17, | |
| "eval_steps_per_second": 3.273, | |
| "eval_wer": 1.0355526473635421, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 21.69, | |
| "eval_loss": 0.7760854959487915, | |
| "eval_runtime": 80.138, | |
| "eval_samples_per_second": 26.142, | |
| "eval_steps_per_second": 3.269, | |
| "eval_wer": 1.0419870221931402, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 22.03, | |
| "learning_rate": 6.0119999999999994e-05, | |
| "loss": 0.0756, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 23.05, | |
| "eval_loss": 0.7345677614212036, | |
| "eval_runtime": 80.4841, | |
| "eval_samples_per_second": 26.03, | |
| "eval_steps_per_second": 3.255, | |
| "eval_wer": 1.036915862369813, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 23.73, | |
| "learning_rate": 4.012e-05, | |
| "loss": 0.0666, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 24.41, | |
| "eval_loss": 0.7506045699119568, | |
| "eval_runtime": 82.6434, | |
| "eval_samples_per_second": 25.35, | |
| "eval_steps_per_second": 3.17, | |
| "eval_wer": 1.0449315666066852, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 25.42, | |
| "learning_rate": 2.0119999999999997e-05, | |
| "loss": 0.0595, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 25.76, | |
| "eval_loss": 0.7319227457046509, | |
| "eval_runtime": 79.8082, | |
| "eval_samples_per_second": 26.25, | |
| "eval_steps_per_second": 3.283, | |
| "eval_wer": 1.0476034680189759, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 27.12, | |
| "learning_rate": 1.6e-07, | |
| "loss": 0.054, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 27.12, | |
| "eval_loss": 0.7346429228782654, | |
| "eval_runtime": 79.9162, | |
| "eval_samples_per_second": 26.215, | |
| "eval_steps_per_second": 3.278, | |
| "eval_wer": 1.0478761110202301, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 27.12, | |
| "step": 8000, | |
| "total_flos": 1.592873144248711e+19, | |
| "train_loss": 0.46973063707351687, | |
| "train_runtime": 10600.4383, | |
| "train_samples_per_second": 12.075, | |
| "train_steps_per_second": 0.755 | |
| } | |
| ], | |
| "max_steps": 8000, | |
| "num_train_epochs": 28, | |
| "total_flos": 1.592873144248711e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |