| { | |
| "best_metric": 31.85479597244303, | |
| "best_model_checkpoint": "./checkpoint-1000", | |
| "epoch": 13.064, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.2e-06, | |
| "loss": 0.9495, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.7e-06, | |
| "loss": 0.2746, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 7.2000000000000005e-06, | |
| "loss": 0.15, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 9.7e-06, | |
| "loss": 0.0876, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 9.755555555555556e-06, | |
| "loss": 0.0795, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 9.466666666666667e-06, | |
| "loss": 0.0743, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 9.188888888888889e-06, | |
| "loss": 0.0431, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 8.91111111111111e-06, | |
| "loss": 0.0413, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "eval_loss": 0.0711669921875, | |
| "eval_runtime": 3657.1623, | |
| "eval_samples_per_second": 0.683, | |
| "eval_steps_per_second": 0.043, | |
| "eval_wer": 36.629570747217805, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 8.633333333333334e-06, | |
| "loss": 0.037, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 8.355555555555556e-06, | |
| "loss": 0.0223, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 8.077777777777778e-06, | |
| "loss": 0.0204, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 7.788888888888889e-06, | |
| "loss": 0.0189, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 7.511111111111111e-06, | |
| "loss": 0.0121, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 7.233333333333334e-06, | |
| "loss": 0.0116, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 6.955555555555557e-06, | |
| "loss": 0.009, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 6.677777777777779e-06, | |
| "loss": 0.006, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "eval_loss": 0.08990478515625, | |
| "eval_runtime": 3646.6012, | |
| "eval_samples_per_second": 0.685, | |
| "eval_steps_per_second": 0.043, | |
| "eval_wer": 33.74668786433492, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "learning_rate": 6.4000000000000006e-06, | |
| "loss": 0.0063, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 6.111111111111112e-06, | |
| "loss": 0.0043, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 5.833333333333334e-06, | |
| "loss": 0.0031, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 5.555555555555557e-06, | |
| "loss": 0.0031, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 5.2777777777777785e-06, | |
| "loss": 0.0021, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 5e-06, | |
| "loss": 0.0016, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "learning_rate": 4.722222222222222e-06, | |
| "loss": 0.0015, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 4.433333333333334e-06, | |
| "loss": 0.0008, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "eval_loss": 0.10394287109375, | |
| "eval_runtime": 3647.236, | |
| "eval_samples_per_second": 0.685, | |
| "eval_steps_per_second": 0.043, | |
| "eval_wer": 32.231054583995764, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "learning_rate": 4.155555555555556e-06, | |
| "loss": 0.0009, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 3.877777777777778e-06, | |
| "loss": 0.0007, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 3.6000000000000003e-06, | |
| "loss": 0.0004, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "learning_rate": 3.322222222222222e-06, | |
| "loss": 0.0004, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 10.01, | |
| "learning_rate": 3.0333333333333337e-06, | |
| "loss": 0.0004, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "learning_rate": 2.755555555555556e-06, | |
| "loss": 0.0003, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 10.05, | |
| "learning_rate": 2.4777777777777782e-06, | |
| "loss": 0.0003, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "learning_rate": 2.2e-06, | |
| "loss": 0.0002, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "eval_loss": 0.1141357421875, | |
| "eval_runtime": 3640.5182, | |
| "eval_samples_per_second": 0.686, | |
| "eval_steps_per_second": 0.043, | |
| "eval_wer": 31.928987811340754, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 1.9222222222222224e-06, | |
| "loss": 0.0001, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 11.06, | |
| "learning_rate": 1.6444444444444447e-06, | |
| "loss": 0.0001, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "learning_rate": 1.3555555555555558e-06, | |
| "loss": 0.0001, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "learning_rate": 1.0777777777777779e-06, | |
| "loss": 0.0001, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 12.06, | |
| "learning_rate": 8.000000000000001e-07, | |
| "loss": 0.0001, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "learning_rate": 5.222222222222223e-07, | |
| "loss": 0.0001, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "learning_rate": 2.444444444444445e-07, | |
| "loss": 0.0001, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 13.06, | |
| "learning_rate": 0.0, | |
| "loss": 0.0001, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 13.06, | |
| "eval_loss": 0.11871337890625, | |
| "eval_runtime": 3861.227, | |
| "eval_samples_per_second": 0.647, | |
| "eval_steps_per_second": 0.041, | |
| "eval_wer": 31.85479597244303, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 13.06, | |
| "step": 1000, | |
| "total_flos": 1.3615895701052457e+20, | |
| "train_loss": 0.04660231500864029, | |
| "train_runtime": 35334.3457, | |
| "train_samples_per_second": 1.811, | |
| "train_steps_per_second": 0.028 | |
| } | |
| ], | |
| "max_steps": 1000, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 1.3615895701052457e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |