| { | |
| "best_metric": 1.0308293530178028, | |
| "best_model_checkpoint": "./whisper-tiny-qlora/checkpoint-600", | |
| "epoch": 0.49170251997541486, | |
| "eval_steps": 200, | |
| "global_step": 800, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.015365703749231715, | |
| "grad_norm": 1.955202579498291, | |
| "learning_rate": 4.800000000000001e-07, | |
| "loss": 2.4438, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03073140749846343, | |
| "grad_norm": 1.465623140335083, | |
| "learning_rate": 9.800000000000001e-07, | |
| "loss": 2.0995, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.046097111247695145, | |
| "grad_norm": 1.489674687385559, | |
| "learning_rate": 1.48e-06, | |
| "loss": 2.4302, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.06146281499692686, | |
| "grad_norm": 1.7222563028335571, | |
| "learning_rate": 1.98e-06, | |
| "loss": 2.2989, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07682851874615858, | |
| "grad_norm": 2.750751256942749, | |
| "learning_rate": 2.4800000000000004e-06, | |
| "loss": 2.2729, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.09219422249539029, | |
| "grad_norm": 2.440593957901001, | |
| "learning_rate": 2.9800000000000003e-06, | |
| "loss": 2.0241, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.107559926244622, | |
| "grad_norm": 2.425014019012451, | |
| "learning_rate": 3.48e-06, | |
| "loss": 2.6992, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.12292562999385372, | |
| "grad_norm": 1.8003255128860474, | |
| "learning_rate": 3.980000000000001e-06, | |
| "loss": 2.0456, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.12292562999385372, | |
| "eval_loss": 2.245432138442993, | |
| "eval_runtime": 51.8391, | |
| "eval_samples_per_second": 7.851, | |
| "eval_steps_per_second": 0.251, | |
| "eval_wer": 1.084672166739036, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.13829133374308544, | |
| "grad_norm": 1.9000691175460815, | |
| "learning_rate": 4.48e-06, | |
| "loss": 2.611, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.15365703749231716, | |
| "grad_norm": 1.4311325550079346, | |
| "learning_rate": 4.960000000000001e-06, | |
| "loss": 2.4044, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.16902274124154887, | |
| "grad_norm": 2.4365923404693604, | |
| "learning_rate": 5.460000000000001e-06, | |
| "loss": 2.1992, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.18438844499078058, | |
| "grad_norm": 2.4958994388580322, | |
| "learning_rate": 5.9600000000000005e-06, | |
| "loss": 2.5628, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.1997541487400123, | |
| "grad_norm": 1.8009036779403687, | |
| "learning_rate": 6.460000000000001e-06, | |
| "loss": 1.8897, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.215119852489244, | |
| "grad_norm": 1.8635932207107544, | |
| "learning_rate": 6.96e-06, | |
| "loss": 2.2831, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.23048555623847572, | |
| "grad_norm": 1.8662347793579102, | |
| "learning_rate": 7.4600000000000006e-06, | |
| "loss": 2.0718, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.24585125998770743, | |
| "grad_norm": 2.4394640922546387, | |
| "learning_rate": 7.960000000000002e-06, | |
| "loss": 2.1359, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.24585125998770743, | |
| "eval_loss": 2.1913249492645264, | |
| "eval_runtime": 49.901, | |
| "eval_samples_per_second": 8.156, | |
| "eval_steps_per_second": 0.261, | |
| "eval_wer": 1.1096396005210596, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.26121696373693915, | |
| "grad_norm": 1.6949491500854492, | |
| "learning_rate": 8.46e-06, | |
| "loss": 2.0452, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.2765826674861709, | |
| "grad_norm": 1.273695468902588, | |
| "learning_rate": 8.96e-06, | |
| "loss": 2.0868, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.29194837123540257, | |
| "grad_norm": 2.592191457748413, | |
| "learning_rate": 9.460000000000001e-06, | |
| "loss": 2.028, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.3073140749846343, | |
| "grad_norm": 2.204875946044922, | |
| "learning_rate": 9.960000000000001e-06, | |
| "loss": 2.3593, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.322679778733866, | |
| "grad_norm": 2.3962159156799316, | |
| "learning_rate": 9.54e-06, | |
| "loss": 2.2455, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.33804548248309774, | |
| "grad_norm": 1.6109391450881958, | |
| "learning_rate": 9.040000000000002e-06, | |
| "loss": 2.1865, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.3534111862323294, | |
| "grad_norm": 1.053389310836792, | |
| "learning_rate": 8.540000000000001e-06, | |
| "loss": 2.0772, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.36877688998156116, | |
| "grad_norm": 4.344364166259766, | |
| "learning_rate": 8.040000000000001e-06, | |
| "loss": 1.7806, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.36877688998156116, | |
| "eval_loss": 2.0955584049224854, | |
| "eval_runtime": 47.7981, | |
| "eval_samples_per_second": 8.515, | |
| "eval_steps_per_second": 0.272, | |
| "eval_wer": 1.0308293530178028, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.38414259373079285, | |
| "grad_norm": 2.7162230014801025, | |
| "learning_rate": 7.540000000000001e-06, | |
| "loss": 1.9152, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.3995082974800246, | |
| "grad_norm": 3.225233793258667, | |
| "learning_rate": 7.04e-06, | |
| "loss": 2.2311, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.4148740012292563, | |
| "grad_norm": 3.202869176864624, | |
| "learning_rate": 6.540000000000001e-06, | |
| "loss": 2.0565, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.430239704978488, | |
| "grad_norm": 2.291895627975464, | |
| "learning_rate": 6.040000000000001e-06, | |
| "loss": 2.0733, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.44560540872771975, | |
| "grad_norm": 2.5822622776031494, | |
| "learning_rate": 5.540000000000001e-06, | |
| "loss": 1.8574, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.46097111247695144, | |
| "grad_norm": 4.14776611328125, | |
| "learning_rate": 5.04e-06, | |
| "loss": 2.0769, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.4763368162261832, | |
| "grad_norm": 4.232518196105957, | |
| "learning_rate": 4.540000000000001e-06, | |
| "loss": 2.1455, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.49170251997541486, | |
| "grad_norm": 4.926309585571289, | |
| "learning_rate": 4.04e-06, | |
| "loss": 2.1011, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.49170251997541486, | |
| "eval_loss": 2.02160906791687, | |
| "eval_runtime": 49.8742, | |
| "eval_samples_per_second": 8.161, | |
| "eval_steps_per_second": 0.261, | |
| "eval_wer": 1.138080764220582, | |
| "step": 800 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 1000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.4677941248e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |