| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 60.0, | |
| "global_step": 1260, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 0.00015, | |
| "loss": 10.7834, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "eval_loss": 3.1524412631988525, | |
| "eval_runtime": 53.4465, | |
| "eval_samples_per_second": 8.719, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 0.0003, | |
| "loss": 3.0788, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "eval_loss": 3.044459104537964, | |
| "eval_runtime": 54.4604, | |
| "eval_samples_per_second": 8.557, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 14.29, | |
| "learning_rate": 0.00023023255813953486, | |
| "loss": 2.9807, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 14.29, | |
| "eval_loss": 2.9658894538879395, | |
| "eval_runtime": 54.1542, | |
| "eval_samples_per_second": 8.605, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 19.05, | |
| "learning_rate": 0.00016046511627906975, | |
| "loss": 2.2974, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 19.05, | |
| "eval_loss": 0.9927442669868469, | |
| "eval_runtime": 55.0174, | |
| "eval_samples_per_second": 8.47, | |
| "eval_wer": 0.9315901489814533, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 23.81, | |
| "learning_rate": 9.069767441860464e-05, | |
| "loss": 0.6448, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 23.81, | |
| "eval_loss": 0.5772319436073303, | |
| "eval_runtime": 54.9367, | |
| "eval_samples_per_second": 8.482, | |
| "eval_wer": 0.6889632107023411, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 28.57, | |
| "learning_rate": 2.0930232558139532e-05, | |
| "loss": 0.3268, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 28.57, | |
| "eval_loss": 0.5437523126602173, | |
| "eval_runtime": 54.9465, | |
| "eval_samples_per_second": 8.481, | |
| "eval_wer": 0.629674673152934, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "learning_rate": 0.00015849056603773584, | |
| "loss": 0.2512, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "eval_loss": 0.5623383522033691, | |
| "eval_runtime": 53.3134, | |
| "eval_samples_per_second": 8.741, | |
| "eval_wer": 0.5916692003648525, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 38.1, | |
| "learning_rate": 0.000130188679245283, | |
| "loss": 0.1599, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 38.1, | |
| "eval_loss": 0.5707418322563171, | |
| "eval_runtime": 54.2873, | |
| "eval_samples_per_second": 8.584, | |
| "eval_wer": 0.5852842809364549, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 42.86, | |
| "learning_rate": 0.00010188679245283017, | |
| "loss": 0.1179, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 42.86, | |
| "eval_loss": 0.5668476223945618, | |
| "eval_runtime": 54.4224, | |
| "eval_samples_per_second": 8.563, | |
| "eval_wer": 0.5655214350866524, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 47.62, | |
| "learning_rate": 7.358490566037735e-05, | |
| "loss": 0.092, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 47.62, | |
| "eval_loss": 0.6057275533676147, | |
| "eval_runtime": 54.8666, | |
| "eval_samples_per_second": 8.493, | |
| "eval_wer": 0.5640012161751292, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 52.38, | |
| "learning_rate": 4.528301886792452e-05, | |
| "loss": 0.0808, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 52.38, | |
| "eval_loss": 0.6176655888557434, | |
| "eval_runtime": 55.0084, | |
| "eval_samples_per_second": 8.471, | |
| "eval_wer": 0.5506232897537245, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 57.14, | |
| "learning_rate": 1.6981132075471696e-05, | |
| "loss": 0.0862, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 57.14, | |
| "eval_loss": 0.639299213886261, | |
| "eval_runtime": 55.7062, | |
| "eval_samples_per_second": 8.365, | |
| "eval_wer": 0.5649133475220431, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "step": 1260, | |
| "total_flos": 1.120772175088509e+19, | |
| "train_runtime": 5433.1006, | |
| "train_samples_per_second": 0.232 | |
| } | |
| ], | |
| "max_steps": 1260, | |
| "num_train_epochs": 60, | |
| "total_flos": 1.120772175088509e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |