| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 29.26829268292683, | |
| "global_step": 7200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 5.7155, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "eval_loss": 3.039255142211914, | |
| "eval_runtime": 56.5805, | |
| "eval_samples_per_second": 8.271, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 0.0002869186046511628, | |
| "loss": 1.6883, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "eval_loss": 0.7244638800621033, | |
| "eval_runtime": 58.2783, | |
| "eval_samples_per_second": 8.03, | |
| "eval_wer": 0.5629911280101394, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "learning_rate": 0.0002694767441860465, | |
| "loss": 0.3776, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "eval_loss": 0.6811308264732361, | |
| "eval_runtime": 57.5618, | |
| "eval_samples_per_second": 8.13, | |
| "eval_wer": 0.49150823827629914, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "learning_rate": 0.00025203488372093025, | |
| "loss": 0.2425, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "eval_loss": 0.629974365234375, | |
| "eval_runtime": 57.7361, | |
| "eval_samples_per_second": 8.106, | |
| "eval_wer": 0.45449936628643856, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 8.13, | |
| "learning_rate": 0.00023459302325581392, | |
| "loss": 0.1872, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 8.13, | |
| "eval_loss": 0.7140640616416931, | |
| "eval_runtime": 58.4584, | |
| "eval_samples_per_second": 8.006, | |
| "eval_wer": 0.44283903675538655, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 9.76, | |
| "learning_rate": 0.00021715116279069765, | |
| "loss": 0.1565, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 9.76, | |
| "eval_loss": 0.7015160322189331, | |
| "eval_runtime": 57.3174, | |
| "eval_samples_per_second": 8.165, | |
| "eval_wer": 0.4385297845373891, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 11.38, | |
| "learning_rate": 0.00019970930232558135, | |
| "loss": 0.1232, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 11.38, | |
| "eval_loss": 0.7360302805900574, | |
| "eval_runtime": 57.5205, | |
| "eval_samples_per_second": 8.136, | |
| "eval_wer": 0.4326996197718631, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "learning_rate": 0.00018226744186046508, | |
| "loss": 0.1131, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "eval_loss": 0.921540379524231, | |
| "eval_runtime": 57.6005, | |
| "eval_samples_per_second": 8.125, | |
| "eval_wer": 0.44714828897338404, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 14.63, | |
| "learning_rate": 0.0001648255813953488, | |
| "loss": 0.1004, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 14.63, | |
| "eval_loss": 0.8807307481765747, | |
| "eval_runtime": 57.8571, | |
| "eval_samples_per_second": 8.089, | |
| "eval_wer": 0.4187579214195184, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 16.26, | |
| "learning_rate": 0.00014738372093023254, | |
| "loss": 0.0851, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 16.26, | |
| "eval_loss": 0.7986534237861633, | |
| "eval_runtime": 57.6046, | |
| "eval_samples_per_second": 8.124, | |
| "eval_wer": 0.43067173637515843, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 17.89, | |
| "learning_rate": 0.00012994186046511627, | |
| "loss": 0.0763, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 17.89, | |
| "eval_loss": 0.8154810667037964, | |
| "eval_runtime": 57.7278, | |
| "eval_samples_per_second": 8.107, | |
| "eval_wer": 0.41698352344740175, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 19.51, | |
| "learning_rate": 0.0001125, | |
| "loss": 0.0664, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 19.51, | |
| "eval_loss": 0.8099461197853088, | |
| "eval_runtime": 57.6128, | |
| "eval_samples_per_second": 8.123, | |
| "eval_wer": 0.4124207858048162, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 21.14, | |
| "learning_rate": 9.505813953488371e-05, | |
| "loss": 0.0611, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 21.14, | |
| "eval_loss": 0.8145996928215027, | |
| "eval_runtime": 57.1919, | |
| "eval_samples_per_second": 8.183, | |
| "eval_wer": 0.4114068441064639, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 22.76, | |
| "learning_rate": 7.761627906976744e-05, | |
| "loss": 0.0573, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 22.76, | |
| "eval_loss": 0.8513393402099609, | |
| "eval_runtime": 58.2107, | |
| "eval_samples_per_second": 8.04, | |
| "eval_wer": 0.4126742712294043, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 24.39, | |
| "learning_rate": 6.0174418604651156e-05, | |
| "loss": 0.0531, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 24.39, | |
| "eval_loss": 0.8830446600914001, | |
| "eval_runtime": 57.6531, | |
| "eval_samples_per_second": 8.118, | |
| "eval_wer": 0.4050697084917617, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 26.02, | |
| "learning_rate": 4.273255813953488e-05, | |
| "loss": 0.0498, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 26.02, | |
| "eval_loss": 0.8851639628410339, | |
| "eval_runtime": 57.3676, | |
| "eval_samples_per_second": 8.158, | |
| "eval_wer": 0.40481622306717363, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 27.64, | |
| "learning_rate": 2.52906976744186e-05, | |
| "loss": 0.0444, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 27.64, | |
| "eval_loss": 0.936349093914032, | |
| "eval_runtime": 57.896, | |
| "eval_samples_per_second": 8.083, | |
| "eval_wer": 0.40887198986058304, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 29.27, | |
| "learning_rate": 7.848837209302325e-06, | |
| "loss": 0.0392, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 29.27, | |
| "eval_loss": 0.9341886043548584, | |
| "eval_runtime": 57.5969, | |
| "eval_samples_per_second": 8.125, | |
| "eval_wer": 0.40126742712294045, | |
| "step": 7200 | |
| } | |
| ], | |
| "max_steps": 7380, | |
| "num_train_epochs": 30, | |
| "total_flos": 4.018481177702621e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |