| { | |
| "best_metric": 98.54998728059017, | |
| "best_model_checkpoint": "./whisper-tiny-en/checkpoint-100", | |
| "epoch": 2.4390243902439024, | |
| "eval_steps": 25, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12195121951219512, | |
| "grad_norm": 60.187870025634766, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 6.9693, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.24390243902439024, | |
| "grad_norm": 51.355934143066406, | |
| "learning_rate": 7e-06, | |
| "loss": 6.7154, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.36585365853658536, | |
| "grad_norm": 25.504573822021484, | |
| "learning_rate": 9.777777777777779e-06, | |
| "loss": 6.0841, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.4878048780487805, | |
| "grad_norm": 21.17405128479004, | |
| "learning_rate": 9.222222222222224e-06, | |
| "loss": 5.6359, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.6097560975609756, | |
| "grad_norm": 17.322952270507812, | |
| "learning_rate": 8.666666666666668e-06, | |
| "loss": 5.2273, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.6097560975609756, | |
| "eval_loss": 4.978211402893066, | |
| "eval_runtime": 36.5444, | |
| "eval_samples_per_second": 4.406, | |
| "eval_steps_per_second": 0.575, | |
| "eval_wer": 101.42457389977105, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.7317073170731707, | |
| "grad_norm": 18.395389556884766, | |
| "learning_rate": 8.111111111111112e-06, | |
| "loss": 4.7246, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.8536585365853658, | |
| "grad_norm": 17.32785987854004, | |
| "learning_rate": 7.555555555555556e-06, | |
| "loss": 4.4016, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.975609756097561, | |
| "grad_norm": 14.042008399963379, | |
| "learning_rate": 7e-06, | |
| "loss": 4.38, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.0975609756097562, | |
| "grad_norm": 14.310500144958496, | |
| "learning_rate": 6.444444444444445e-06, | |
| "loss": 4.0846, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.2195121951219512, | |
| "grad_norm": 15.815790176391602, | |
| "learning_rate": 5.88888888888889e-06, | |
| "loss": 4.0984, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.2195121951219512, | |
| "eval_loss": 4.143329620361328, | |
| "eval_runtime": 56.3975, | |
| "eval_samples_per_second": 2.855, | |
| "eval_steps_per_second": 0.372, | |
| "eval_wer": 100.8903586873569, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.3414634146341464, | |
| "grad_norm": 14.383249282836914, | |
| "learning_rate": 5.333333333333334e-06, | |
| "loss": 3.8944, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.4634146341463414, | |
| "grad_norm": 13.232712745666504, | |
| "learning_rate": 4.777777777777778e-06, | |
| "loss": 3.9257, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.5853658536585367, | |
| "grad_norm": 13.823458671569824, | |
| "learning_rate": 4.222222222222223e-06, | |
| "loss": 3.7674, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.7073170731707317, | |
| "grad_norm": 14.021493911743164, | |
| "learning_rate": 3.6666666666666666e-06, | |
| "loss": 3.7766, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.8292682926829267, | |
| "grad_norm": 12.970163345336914, | |
| "learning_rate": 3.1111111111111116e-06, | |
| "loss": 3.8301, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.8292682926829267, | |
| "eval_loss": 3.915684223175049, | |
| "eval_runtime": 56.3392, | |
| "eval_samples_per_second": 2.858, | |
| "eval_steps_per_second": 0.373, | |
| "eval_wer": 99.31315186975324, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.951219512195122, | |
| "grad_norm": 12.832784652709961, | |
| "learning_rate": 2.5555555555555557e-06, | |
| "loss": 3.7075, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.073170731707317, | |
| "grad_norm": 16.47066879272461, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 3.7456, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.1951219512195124, | |
| "grad_norm": 13.32690715789795, | |
| "learning_rate": 1.4444444444444445e-06, | |
| "loss": 3.7669, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.317073170731707, | |
| "grad_norm": 13.807867050170898, | |
| "learning_rate": 8.88888888888889e-07, | |
| "loss": 3.5748, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 2.4390243902439024, | |
| "grad_norm": 12.753012657165527, | |
| "learning_rate": 3.3333333333333335e-07, | |
| "loss": 3.7081, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.4390243902439024, | |
| "eval_loss": 3.854832887649536, | |
| "eval_runtime": 56.4689, | |
| "eval_samples_per_second": 2.851, | |
| "eval_steps_per_second": 0.372, | |
| "eval_wer": 98.54998728059017, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.875012517888e+16, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |