| { |
| "best_global_step": 18000, |
| "best_metric": 28.542471111364886, |
| "best_model_checkpoint": "./whisper-small-yoruba-438h/checkpoint-18000", |
| "epoch": 6.581352833638025, |
| "eval_steps": 2000, |
| "global_step": 18000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.18281535648994515, |
| "grad_norm": 23.418228149414062, |
| "learning_rate": 2.4950000000000003e-06, |
| "loss": 8.61669921875, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.3656307129798903, |
| "grad_norm": 18.097488403320312, |
| "learning_rate": 4.9950000000000005e-06, |
| "loss": 2.76252197265625, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.5484460694698354, |
| "grad_norm": 14.418298721313477, |
| "learning_rate": 4.980903172380229e-06, |
| "loss": 1.8946580810546876, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.7312614259597806, |
| "grad_norm": 13.605561256408691, |
| "learning_rate": 4.9237526483866514e-06, |
| "loss": 1.6098297119140625, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.7312614259597806, |
| "eval_loss": 0.37521448731422424, |
| "eval_runtime": 6135.7464, |
| "eval_samples_per_second": 1.585, |
| "eval_steps_per_second": 0.198, |
| "eval_wer": 42.951101658018516, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.9140767824497258, |
| "grad_norm": 13.122516632080078, |
| "learning_rate": 4.829424964672522e-06, |
| "loss": 1.4408970947265625, |
| "step": 2500 |
| }, |
| { |
| "epoch": 1.0968921389396709, |
| "grad_norm": 11.27369499206543, |
| "learning_rate": 4.699366983138475e-06, |
| "loss": 1.28490234375, |
| "step": 3000 |
| }, |
| { |
| "epoch": 1.2797074954296161, |
| "grad_norm": 12.354094505310059, |
| "learning_rate": 4.535573621193117e-06, |
| "loss": 1.175876220703125, |
| "step": 3500 |
| }, |
| { |
| "epoch": 1.4625228519195612, |
| "grad_norm": 12.436117172241211, |
| "learning_rate": 4.340557252358663e-06, |
| "loss": 1.1245360107421876, |
| "step": 4000 |
| }, |
| { |
| "epoch": 1.4625228519195612, |
| "eval_loss": 0.29289114475250244, |
| "eval_runtime": 4432.1972, |
| "eval_samples_per_second": 2.194, |
| "eval_steps_per_second": 0.274, |
| "eval_wer": 36.84005795570528, |
| "step": 4000 |
| }, |
| { |
| "epoch": 1.6453382084095063, |
| "grad_norm": 11.27859115600586, |
| "learning_rate": 4.1173091697842365e-06, |
| "loss": 1.087154541015625, |
| "step": 4500 |
| }, |
| { |
| "epoch": 1.8281535648994516, |
| "grad_norm": 12.39543628692627, |
| "learning_rate": 3.869253703765544e-06, |
| "loss": 1.0696717529296875, |
| "step": 5000 |
| }, |
| { |
| "epoch": 2.010968921389397, |
| "grad_norm": 10.303967475891113, |
| "learning_rate": 3.6001956970475217e-06, |
| "loss": 1.0127244873046874, |
| "step": 5500 |
| }, |
| { |
| "epoch": 2.1937842778793417, |
| "grad_norm": 11.98684024810791, |
| "learning_rate": 3.3142621435693033e-06, |
| "loss": 0.8952516479492187, |
| "step": 6000 |
| }, |
| { |
| "epoch": 2.1937842778793417, |
| "eval_loss": 0.26017212867736816, |
| "eval_runtime": 4412.3584, |
| "eval_samples_per_second": 2.204, |
| "eval_steps_per_second": 0.276, |
| "eval_wer": 33.975104741376235, |
| "step": 6000 |
| }, |
| { |
| "epoch": 2.376599634369287, |
| "grad_norm": 9.57399845123291, |
| "learning_rate": 4.481673449215052e-06, |
| "loss": 0.8934461059570312, |
| "step": 6500 |
| }, |
| { |
| "epoch": 2.5594149908592323, |
| "grad_norm": 12.205296516418457, |
| "learning_rate": 4.387349612915318e-06, |
| "loss": 0.8670662231445313, |
| "step": 7000 |
| }, |
| { |
| "epoch": 2.742230347349177, |
| "grad_norm": 11.39163875579834, |
| "learning_rate": 4.286320722991205e-06, |
| "loss": 0.86044384765625, |
| "step": 7500 |
| }, |
| { |
| "epoch": 2.9250457038391224, |
| "grad_norm": 11.357952117919922, |
| "learning_rate": 4.178945697642034e-06, |
| "loss": 0.8341318969726562, |
| "step": 8000 |
| }, |
| { |
| "epoch": 2.9250457038391224, |
| "eval_loss": 0.24125488102436066, |
| "eval_runtime": 4321.5359, |
| "eval_samples_per_second": 2.25, |
| "eval_steps_per_second": 0.281, |
| "eval_wer": 32.54798118580799, |
| "step": 8000 |
| }, |
| { |
| "epoch": 3.1078610603290677, |
| "grad_norm": 12.443680763244629, |
| "learning_rate": 4.0656060005340766e-06, |
| "loss": 0.7627076416015625, |
| "step": 8500 |
| }, |
| { |
| "epoch": 3.2906764168190126, |
| "grad_norm": 10.662081718444824, |
| "learning_rate": 3.946704285601551e-06, |
| "loss": 0.7079920043945312, |
| "step": 9000 |
| }, |
| { |
| "epoch": 3.473491773308958, |
| "grad_norm": 13.675153732299805, |
| "learning_rate": 3.822662966566427e-06, |
| "loss": 0.7093059692382813, |
| "step": 9500 |
| }, |
| { |
| "epoch": 3.656307129798903, |
| "grad_norm": 11.558106422424316, |
| "learning_rate": 3.6939227162590474e-06, |
| "loss": 0.699214599609375, |
| "step": 10000 |
| }, |
| { |
| "epoch": 3.656307129798903, |
| "eval_loss": 0.22658970952033997, |
| "eval_runtime": 4394.9833, |
| "eval_samples_per_second": 2.213, |
| "eval_steps_per_second": 0.277, |
| "eval_wer": 31.3628155623916, |
| "step": 10000 |
| }, |
| { |
| "epoch": 3.839122486288848, |
| "grad_norm": 12.224128723144531, |
| "learning_rate": 3.5609409010708955e-06, |
| "loss": 0.7020602416992188, |
| "step": 10500 |
| }, |
| { |
| "epoch": 4.021937842778794, |
| "grad_norm": 13.084427833557129, |
| "learning_rate": 3.4241899561013316e-06, |
| "loss": 0.6613292846679687, |
| "step": 11000 |
| }, |
| { |
| "epoch": 4.204753199268739, |
| "grad_norm": 10.680255889892578, |
| "learning_rate": 3.284155706770791e-06, |
| "loss": 0.5683858032226563, |
| "step": 11500 |
| }, |
| { |
| "epoch": 4.387568555758683, |
| "grad_norm": 12.943653106689453, |
| "learning_rate": 3.1413356428631416e-06, |
| "loss": 0.5714318237304687, |
| "step": 12000 |
| }, |
| { |
| "epoch": 4.387568555758683, |
| "eval_loss": 0.21931344270706177, |
| "eval_runtime": 4368.7783, |
| "eval_samples_per_second": 2.226, |
| "eval_steps_per_second": 0.278, |
| "eval_wer": 29.561335265190174, |
| "step": 12000 |
| }, |
| { |
| "epoch": 4.570383912248629, |
| "grad_norm": 9.87512493133545, |
| "learning_rate": 2.996237151128874e-06, |
| "loss": 0.5676758422851562, |
| "step": 12500 |
| }, |
| { |
| "epoch": 4.753199268738574, |
| "grad_norm": 11.12316608428955, |
| "learning_rate": 2.8493757127280436e-06, |
| "loss": 0.5684054565429687, |
| "step": 13000 |
| }, |
| { |
| "epoch": 4.936014625228519, |
| "grad_norm": 10.787373542785645, |
| "learning_rate": 2.701273071916772e-06, |
| "loss": 0.5666332397460937, |
| "step": 13500 |
| }, |
| { |
| "epoch": 5.118829981718465, |
| "grad_norm": 10.713043212890625, |
| "learning_rate": 2.552455382483289e-06, |
| "loss": 0.4890457763671875, |
| "step": 14000 |
| }, |
| { |
| "epoch": 5.118829981718465, |
| "eval_loss": 0.2142162322998047, |
| "eval_runtime": 4395.2429, |
| "eval_samples_per_second": 2.213, |
| "eval_steps_per_second": 0.277, |
| "eval_wer": 29.108467100144892, |
| "step": 14000 |
| }, |
| { |
| "epoch": 5.301645338208409, |
| "grad_norm": 10.574690818786621, |
| "learning_rate": 2.403451338518546e-06, |
| "loss": 0.46753656005859373, |
| "step": 14500 |
| }, |
| { |
| "epoch": 5.484460694698354, |
| "grad_norm": 12.801483154296875, |
| "learning_rate": 2.2547902961620725e-06, |
| "loss": 0.4668481140136719, |
| "step": 15000 |
| }, |
| { |
| "epoch": 5.6672760511883, |
| "grad_norm": 12.052448272705078, |
| "learning_rate": 2.107000392995815e-06, |
| "loss": 0.46311260986328123, |
| "step": 15500 |
| }, |
| { |
| "epoch": 5.850091407678245, |
| "grad_norm": 11.316293716430664, |
| "learning_rate": 1.960606671767071e-06, |
| "loss": 0.45828167724609375, |
| "step": 16000 |
| }, |
| { |
| "epoch": 5.850091407678245, |
| "eval_loss": 0.2103041559457779, |
| "eval_runtime": 4422.576, |
| "eval_samples_per_second": 2.199, |
| "eval_steps_per_second": 0.275, |
| "eval_wer": 29.040304909818925, |
| "step": 16000 |
| }, |
| { |
| "epoch": 6.03290676416819, |
| "grad_norm": 8.880300521850586, |
| "learning_rate": 1.8161292151062066e-06, |
| "loss": 0.44292330932617185, |
| "step": 16500 |
| }, |
| { |
| "epoch": 6.2157221206581355, |
| "grad_norm": 8.955211639404297, |
| "learning_rate": 1.6740812978658517e-06, |
| "loss": 0.38158233642578127, |
| "step": 17000 |
| }, |
| { |
| "epoch": 6.39853747714808, |
| "grad_norm": 11.817580223083496, |
| "learning_rate": 1.5349675636455847e-06, |
| "loss": 0.3822865295410156, |
| "step": 17500 |
| }, |
| { |
| "epoch": 6.581352833638025, |
| "grad_norm": 10.142769813537598, |
| "learning_rate": 1.3992822319802764e-06, |
| "loss": 0.38730328369140626, |
| "step": 18000 |
| }, |
| { |
| "epoch": 6.581352833638025, |
| "eval_loss": 0.2117231786251068, |
| "eval_runtime": 4368.2868, |
| "eval_samples_per_second": 2.226, |
| "eval_steps_per_second": 0.278, |
| "eval_wer": 28.542471111364886, |
| "step": 18000 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 27350, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 10, |
| "save_steps": 2000, |
| "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": 1.6621999604711424e+20, |
| "train_batch_size": 8, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
