| { | |
| "best_metric": 111.13240084307338, | |
| "best_model_checkpoint": "./whisper-tiny-ha/checkpoint-1000", | |
| "epoch": 6.369426751592357, | |
| "eval_steps": 500, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1592356687898089, | |
| "grad_norm": 30.159414291381836, | |
| "learning_rate": 4.2000000000000004e-06, | |
| "loss": 3.418, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.3184713375796178, | |
| "grad_norm": 21.70500373840332, | |
| "learning_rate": 9.200000000000002e-06, | |
| "loss": 2.553, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.47770700636942676, | |
| "grad_norm": 18.039278030395508, | |
| "learning_rate": 1e-05, | |
| "loss": 2.3832, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.6369426751592356, | |
| "grad_norm": 18.557083129882812, | |
| "learning_rate": 1e-05, | |
| "loss": 2.3115, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.7961783439490446, | |
| "grad_norm": 16.37763023376465, | |
| "learning_rate": 1e-05, | |
| "loss": 2.2126, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.9554140127388535, | |
| "grad_norm": 18.549989700317383, | |
| "learning_rate": 1e-05, | |
| "loss": 2.1729, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.1146496815286624, | |
| "grad_norm": 16.92412757873535, | |
| "learning_rate": 1e-05, | |
| "loss": 1.9976, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.2738853503184713, | |
| "grad_norm": 17.25154685974121, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8783, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.4331210191082802, | |
| "grad_norm": 16.648714065551758, | |
| "learning_rate": 1e-05, | |
| "loss": 1.902, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.5923566878980893, | |
| "grad_norm": 16.841991424560547, | |
| "learning_rate": 1e-05, | |
| "loss": 1.925, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.7515923566878981, | |
| "grad_norm": 19.01357650756836, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8552, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.910828025477707, | |
| "grad_norm": 16.73857879638672, | |
| "learning_rate": 1e-05, | |
| "loss": 1.8734, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.070063694267516, | |
| "grad_norm": 16.232080459594727, | |
| "learning_rate": 1e-05, | |
| "loss": 1.756, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 2.229299363057325, | |
| "grad_norm": 16.035682678222656, | |
| "learning_rate": 1e-05, | |
| "loss": 1.584, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.388535031847134, | |
| "grad_norm": 15.77041244506836, | |
| "learning_rate": 1e-05, | |
| "loss": 1.5446, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.5477707006369426, | |
| "grad_norm": 17.873939514160156, | |
| "learning_rate": 1e-05, | |
| "loss": 1.5419, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.7070063694267517, | |
| "grad_norm": 16.445964813232422, | |
| "learning_rate": 1e-05, | |
| "loss": 1.6085, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.8662420382165603, | |
| "grad_norm": 16.27969741821289, | |
| "learning_rate": 1e-05, | |
| "loss": 1.6325, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.0254777070063694, | |
| "grad_norm": 15.765183448791504, | |
| "learning_rate": 1e-05, | |
| "loss": 1.4205, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 3.1847133757961785, | |
| "grad_norm": 14.563206672668457, | |
| "learning_rate": 1e-05, | |
| "loss": 1.3176, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.1847133757961785, | |
| "eval_loss": 2.107278823852539, | |
| "eval_runtime": 154.4492, | |
| "eval_samples_per_second": 4.273, | |
| "eval_steps_per_second": 0.272, | |
| "eval_wer": 132.4391645909178, | |
| "eval_wer_ortho": 133.80859375, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.343949044585987, | |
| "grad_norm": 18.346054077148438, | |
| "learning_rate": 1e-05, | |
| "loss": 1.3377, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 3.5031847133757963, | |
| "grad_norm": 16.357585906982422, | |
| "learning_rate": 1e-05, | |
| "loss": 1.3085, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.662420382165605, | |
| "grad_norm": 16.26805305480957, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2712, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 3.821656050955414, | |
| "grad_norm": 16.663305282592773, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2537, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.980891719745223, | |
| "grad_norm": 15.779778480529785, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2877, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 4.140127388535032, | |
| "grad_norm": 14.90344524383545, | |
| "learning_rate": 1e-05, | |
| "loss": 1.1309, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 4.2993630573248405, | |
| "grad_norm": 15.136467933654785, | |
| "learning_rate": 1e-05, | |
| "loss": 1.0086, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 4.45859872611465, | |
| "grad_norm": 16.454153060913086, | |
| "learning_rate": 1e-05, | |
| "loss": 1.0473, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.617834394904459, | |
| "grad_norm": 17.21575927734375, | |
| "learning_rate": 1e-05, | |
| "loss": 1.0229, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 4.777070063694268, | |
| "grad_norm": 18.05030059814453, | |
| "learning_rate": 1e-05, | |
| "loss": 1.0785, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 4.936305732484076, | |
| "grad_norm": 17.032936096191406, | |
| "learning_rate": 1e-05, | |
| "loss": 1.0569, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 5.095541401273885, | |
| "grad_norm": 16.049070358276367, | |
| "learning_rate": 1e-05, | |
| "loss": 0.9171, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.254777070063694, | |
| "grad_norm": 15.811676025390625, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8208, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 5.414012738853503, | |
| "grad_norm": 15.85254955291748, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8272, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 5.573248407643312, | |
| "grad_norm": 17.296329498291016, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8312, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 5.732484076433121, | |
| "grad_norm": 16.399415969848633, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8444, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.89171974522293, | |
| "grad_norm": 16.311349868774414, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8216, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 6.050955414012739, | |
| "grad_norm": 14.990595817565918, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7495, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 6.210191082802548, | |
| "grad_norm": 13.810046195983887, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5933, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 6.369426751592357, | |
| "grad_norm": 12.30604362487793, | |
| "learning_rate": 1e-05, | |
| "loss": 0.624, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.369426751592357, | |
| "eval_loss": 2.2332684993743896, | |
| "eval_runtime": 146.206, | |
| "eval_samples_per_second": 4.514, | |
| "eval_steps_per_second": 0.287, | |
| "eval_wer": 111.13240084307338, | |
| "eval_wer_ortho": 110.44921875, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 2355, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "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": 3.9331130867712e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |