| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.660377358490566, | |
| "global_step": 1800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 1.94e-05, | |
| "loss": 29.5423, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "eval_loss": 3.126616954803467, | |
| "eval_runtime": 93.155, | |
| "eval_samples_per_second": 9.125, | |
| "eval_steps_per_second": 1.149, | |
| "eval_wer": 0.9997463216641299, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 3.94e-05, | |
| "loss": 3.1063, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "eval_loss": 3.0489962100982666, | |
| "eval_runtime": 93.7928, | |
| "eval_samples_per_second": 9.063, | |
| "eval_steps_per_second": 1.141, | |
| "eval_wer": 0.9997463216641299, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 5.94e-05, | |
| "loss": 3.0009, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_loss": 3.005607843399048, | |
| "eval_runtime": 92.9047, | |
| "eval_samples_per_second": 9.149, | |
| "eval_steps_per_second": 1.152, | |
| "eval_wer": 0.9997463216641299, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 7.94e-05, | |
| "loss": 2.6697, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "eval_loss": 0.6328503489494324, | |
| "eval_runtime": 93.7379, | |
| "eval_samples_per_second": 9.068, | |
| "eval_steps_per_second": 1.141, | |
| "eval_wer": 0.3745983426348723, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 9.94e-05, | |
| "loss": 0.6952, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "eval_loss": 0.4102373421192169, | |
| "eval_runtime": 95.3762, | |
| "eval_samples_per_second": 8.912, | |
| "eval_steps_per_second": 1.122, | |
| "eval_wer": 0.27354980551327585, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 9.589331075359865e-05, | |
| "loss": 0.6951, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "eval_loss": 0.3678465485572815, | |
| "eval_runtime": 93.5395, | |
| "eval_samples_per_second": 9.087, | |
| "eval_steps_per_second": 1.144, | |
| "eval_wer": 0.2611195670556401, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "learning_rate": 9.165961049957663e-05, | |
| "loss": 0.4454, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "eval_loss": 0.37026697397232056, | |
| "eval_runtime": 94.3783, | |
| "eval_samples_per_second": 9.006, | |
| "eval_steps_per_second": 1.134, | |
| "eval_wer": 0.24801285303568407, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 8.742591024555461e-05, | |
| "loss": 0.5078, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "eval_loss": 0.3470672369003296, | |
| "eval_runtime": 93.8587, | |
| "eval_samples_per_second": 9.056, | |
| "eval_steps_per_second": 1.14, | |
| "eval_wer": 0.24708269913749364, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 8.31922099915326e-05, | |
| "loss": 0.3257, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "eval_loss": 0.33427906036376953, | |
| "eval_runtime": 94.4723, | |
| "eval_samples_per_second": 8.997, | |
| "eval_steps_per_second": 1.133, | |
| "eval_wer": 0.2371892440385591, | |
| "step": 1800 | |
| } | |
| ], | |
| "max_steps": 5724, | |
| "num_train_epochs": 18, | |
| "total_flos": 1.411511646744576e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |