| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "global_step": 6250, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.000276096, | |
| "loss": 2.6773, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 1.390709400177002, | |
| "eval_runtime": 221.8176, | |
| "eval_samples_per_second": 22.541, | |
| "eval_steps_per_second": 2.818, | |
| "eval_wer": 0.9864438767290387, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.00025209599999999994, | |
| "loss": 0.9526, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 0.7760354280471802, | |
| "eval_runtime": 219.0168, | |
| "eval_samples_per_second": 22.829, | |
| "eval_steps_per_second": 2.854, | |
| "eval_wer": 0.644847982692352, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.00022809599999999998, | |
| "loss": 0.6418, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_loss": 0.760543942451477, | |
| "eval_runtime": 217.3821, | |
| "eval_samples_per_second": 23.001, | |
| "eval_steps_per_second": 2.875, | |
| "eval_wer": 0.6193928513889848, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 0.000204096, | |
| "loss": 0.5028, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "eval_loss": 0.6515631675720215, | |
| "eval_runtime": 217.5138, | |
| "eval_samples_per_second": 22.987, | |
| "eval_steps_per_second": 2.873, | |
| "eval_wer": 0.532187161959999, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.00018009599999999998, | |
| "loss": 0.4133, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.630308210849762, | |
| "eval_runtime": 220.9761, | |
| "eval_samples_per_second": 22.627, | |
| "eval_steps_per_second": 2.828, | |
| "eval_wer": 0.5097010287923772, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 0.000156096, | |
| "loss": 0.3285, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "eval_loss": 0.6422050595283508, | |
| "eval_runtime": 216.2337, | |
| "eval_samples_per_second": 23.123, | |
| "eval_steps_per_second": 2.89, | |
| "eval_wer": 0.5062487053787199, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 0.00013209599999999998, | |
| "loss": 0.2764, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "eval_loss": 0.5936103463172913, | |
| "eval_runtime": 221.9715, | |
| "eval_samples_per_second": 22.525, | |
| "eval_steps_per_second": 2.816, | |
| "eval_wer": 0.47476351584616444, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 0.00010809599999999998, | |
| "loss": 0.2361, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "eval_loss": 0.6486021280288696, | |
| "eval_runtime": 217.2436, | |
| "eval_samples_per_second": 23.016, | |
| "eval_steps_per_second": 2.877, | |
| "eval_wer": 0.4683191788073373, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 8.4144e-05, | |
| "loss": 0.2049, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "eval_loss": 0.6320939064025879, | |
| "eval_runtime": 216.2222, | |
| "eval_samples_per_second": 23.124, | |
| "eval_steps_per_second": 2.891, | |
| "eval_wer": 0.453198002255518, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 6.014399999999999e-05, | |
| "loss": 0.176, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.6229676604270935, | |
| "eval_runtime": 217.7281, | |
| "eval_samples_per_second": 22.964, | |
| "eval_steps_per_second": 2.871, | |
| "eval_wer": 0.44822665653985133, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 8.8, | |
| "learning_rate": 3.6191999999999995e-05, | |
| "loss": 0.1393, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 8.8, | |
| "eval_loss": 0.6595410108566284, | |
| "eval_runtime": 217.4753, | |
| "eval_samples_per_second": 22.991, | |
| "eval_steps_per_second": 2.874, | |
| "eval_wer": 0.4402863126884393, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "learning_rate": 1.2192e-05, | |
| "loss": 0.1141, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "eval_loss": 0.6552333831787109, | |
| "eval_runtime": 216.5487, | |
| "eval_samples_per_second": 23.089, | |
| "eval_steps_per_second": 2.886, | |
| "eval_wer": 0.43480862620543625, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 6250, | |
| "total_flos": 1.0733519285960786e+19, | |
| "train_loss": 0.5376352182006836, | |
| "train_runtime": 6902.5833, | |
| "train_samples_per_second": 14.487, | |
| "train_steps_per_second": 0.905 | |
| } | |
| ], | |
| "max_steps": 6250, | |
| "num_train_epochs": 10, | |
| "total_flos": 1.0733519285960786e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |