| { | |
| "best_metric": 100.24908986395862, | |
| "best_model_checkpoint": "./whisper-small-ha/checkpoint-500", | |
| "epoch": 3.1847133757961785, | |
| "eval_steps": 500, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1592356687898089, | |
| "grad_norm": 1.5306243896484375, | |
| "learning_rate": 5e-06, | |
| "loss": 0.0034, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.3184713375796178, | |
| "grad_norm": 0.7897353172302246, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0053, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.47770700636942676, | |
| "grad_norm": 0.6488688588142395, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0075, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.6369426751592356, | |
| "grad_norm": 5.7697882652282715, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0117, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.7961783439490446, | |
| "grad_norm": 3.655069351196289, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0112, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.9554140127388535, | |
| "grad_norm": 0.6484639644622803, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0119, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.1146496815286624, | |
| "grad_norm": 1.6433720588684082, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0078, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.2738853503184713, | |
| "grad_norm": 0.5470134019851685, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0111, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.4331210191082802, | |
| "grad_norm": 7.9058685302734375, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0105, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.5923566878980893, | |
| "grad_norm": 0.44242119789123535, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0083, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.7515923566878981, | |
| "grad_norm": 0.560337483882904, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0093, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.910828025477707, | |
| "grad_norm": 0.45471709966659546, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0065, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.070063694267516, | |
| "grad_norm": 0.8609573841094971, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0071, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 2.229299363057325, | |
| "grad_norm": 0.27363401651382446, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0044, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.388535031847134, | |
| "grad_norm": 0.8000548481941223, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0057, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.5477707006369426, | |
| "grad_norm": 13.804436683654785, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0067, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.7070063694267517, | |
| "grad_norm": 0.20495767891407013, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0064, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.8662420382165603, | |
| "grad_norm": 1.7401798963546753, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0097, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.0254777070063694, | |
| "grad_norm": 2.01271653175354, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0067, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 3.1847133757961785, | |
| "grad_norm": 0.6848011016845703, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0081, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.1847133757961785, | |
| "eval_loss": 2.595545768737793, | |
| "eval_runtime": 162.5603, | |
| "eval_samples_per_second": 4.06, | |
| "eval_steps_per_second": 0.258, | |
| "eval_wer": 100.24908986395862, | |
| "eval_wer_ortho": 101.58203125, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 3611, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 23, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.1810164932608e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |