| { | |
| "best_metric": 98.29925818708159, | |
| "best_model_checkpoint": "./whisper-bangla-dialect/checkpoint-1620", | |
| "epoch": 4.992295839753467, | |
| "eval_steps": 500, | |
| "global_step": 1620, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15408320493066255, | |
| "grad_norm": 4.895554065704346, | |
| "learning_rate": 4.9000000000000005e-06, | |
| "loss": 1.7624, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.3081664098613251, | |
| "grad_norm": 3.634995460510254, | |
| "learning_rate": 9.9e-06, | |
| "loss": 1.5119, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4622496147919877, | |
| "grad_norm": 2.232389211654663, | |
| "learning_rate": 1.4900000000000001e-05, | |
| "loss": 1.4947, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.6163328197226502, | |
| "grad_norm": 2.1193408966064453, | |
| "learning_rate": 1.9900000000000003e-05, | |
| "loss": 1.473, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7704160246533128, | |
| "grad_norm": 1.3372803926467896, | |
| "learning_rate": 1.9941297059390586e-05, | |
| "loss": 1.4826, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.9244992295839753, | |
| "grad_norm": 3.3661787509918213, | |
| "learning_rate": 1.97610942687054e-05, | |
| "loss": 1.4619, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9984591679506933, | |
| "eval_loss": 1.4470419883728027, | |
| "eval_norm_levenshtein_similarity": 95.46770399855255, | |
| "eval_runtime": 351.6357, | |
| "eval_samples_per_second": 1.641, | |
| "eval_steps_per_second": 0.412, | |
| "eval_wer": 5.913978494623656, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 1.078582434514638, | |
| "grad_norm": 1.7728036642074585, | |
| "learning_rate": 1.946156998812636e-05, | |
| "loss": 1.457, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.2326656394453004, | |
| "grad_norm": 2.1206459999084473, | |
| "learning_rate": 1.9046385659950213e-05, | |
| "loss": 1.4547, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.386748844375963, | |
| "grad_norm": 1.305923342704773, | |
| "learning_rate": 1.8520616577095447e-05, | |
| "loss": 1.4502, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.5408320493066254, | |
| "grad_norm": 3.612596035003662, | |
| "learning_rate": 1.7890689841747463e-05, | |
| "loss": 1.4495, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.694915254237288, | |
| "grad_norm": 1.3930879831314087, | |
| "learning_rate": 1.716430579922786e-05, | |
| "loss": 1.442, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.8489984591679507, | |
| "grad_norm": 1.770542025566101, | |
| "learning_rate": 1.6350343907495195e-05, | |
| "loss": 1.4445, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.434973120689392, | |
| "eval_norm_levenshtein_similarity": 96.86991134430976, | |
| "eval_runtime": 348.1264, | |
| "eval_samples_per_second": 1.657, | |
| "eval_steps_per_second": 0.417, | |
| "eval_wer": 4.69208211143695, | |
| "step": 649 | |
| }, | |
| { | |
| "epoch": 2.0030816640986133, | |
| "grad_norm": 1.691462755203247, | |
| "learning_rate": 1.54587541929466e-05, | |
| "loss": 1.4416, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.157164869029276, | |
| "grad_norm": 1.5759495496749878, | |
| "learning_rate": 1.4500435619385243e-05, | |
| "loss": 1.4332, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.3112480739599386, | |
| "grad_norm": 1.881423830986023, | |
| "learning_rate": 1.348710285699486e-05, | |
| "loss": 1.4319, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.4653312788906008, | |
| "grad_norm": 1.305807113647461, | |
| "learning_rate": 1.2431143079962917e-05, | |
| "loss": 1.4324, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.6194144838212634, | |
| "grad_norm": 1.3964393138885498, | |
| "learning_rate": 1.1345464543285721e-05, | |
| "loss": 1.4343, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.773497688751926, | |
| "grad_norm": 1.2005863189697266, | |
| "learning_rate": 1.0243338789781722e-05, | |
| "loss": 1.431, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.9275808936825887, | |
| "grad_norm": 0.7941581606864929, | |
| "learning_rate": 9.13823841620477e-06, | |
| "loss": 1.432, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.9984591679506933, | |
| "eval_loss": 1.427686095237732, | |
| "eval_norm_levenshtein_similarity": 97.37651528858332, | |
| "eval_runtime": 346.2984, | |
| "eval_samples_per_second": 1.666, | |
| "eval_steps_per_second": 0.419, | |
| "eval_wer": 3.5679374389051812, | |
| "step": 973 | |
| }, | |
| { | |
| "epoch": 3.0816640986132513, | |
| "grad_norm": 0.9101853966712952, | |
| "learning_rate": 8.043672381635423e-06, | |
| "loss": 1.4277, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.2357473035439135, | |
| "grad_norm": 0.7756341695785522, | |
| "learning_rate": 6.9730208713722456e-06, | |
| "loss": 1.425, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.389830508474576, | |
| "grad_norm": 1.314003586769104, | |
| "learning_rate": 5.9393717349785064e-06, | |
| "loss": 1.4231, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.5439137134052388, | |
| "grad_norm": 2.220507860183716, | |
| "learning_rate": 4.955360497897032e-06, | |
| "loss": 1.4263, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 3.6979969183359014, | |
| "grad_norm": 0.39077863097190857, | |
| "learning_rate": 4.033015902361899e-06, | |
| "loss": 1.423, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.852080123266564, | |
| "grad_norm": 0.3764391839504242, | |
| "learning_rate": 3.183612865744584e-06, | |
| "loss": 1.4232, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.4236760139465332, | |
| "eval_norm_levenshtein_similarity": 98.16356070200833, | |
| "eval_runtime": 346.601, | |
| "eval_samples_per_second": 1.665, | |
| "eval_steps_per_second": 0.418, | |
| "eval_wer": 2.3949169110459434, | |
| "step": 1298 | |
| }, | |
| { | |
| "epoch": 4.006163328197227, | |
| "grad_norm": 0.7122958898544312, | |
| "learning_rate": 2.4175346537998824e-06, | |
| "loss": 1.4221, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.160246533127889, | |
| "grad_norm": 0.48986372351646423, | |
| "learning_rate": 1.744145953632339e-06, | |
| "loss": 1.4217, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 4.314329738058552, | |
| "grad_norm": 1.2581398487091064, | |
| "learning_rate": 1.1716783979636393e-06, | |
| "loss": 1.4206, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.4684129429892145, | |
| "grad_norm": 0.4906635880470276, | |
| "learning_rate": 7.071299400742126e-07, | |
| "loss": 1.4204, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 4.622496147919877, | |
| "grad_norm": 0.4289968013763428, | |
| "learning_rate": 3.56179309479151e-07, | |
| "loss": 1.4199, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.776579352850539, | |
| "grad_norm": 0.8411958813667297, | |
| "learning_rate": 1.231165940486234e-07, | |
| "loss": 1.4201, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 4.9306625577812015, | |
| "grad_norm": 0.30201002955436707, | |
| "learning_rate": 1.079079715032183e-08, | |
| "loss": 1.422, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.992295839753467, | |
| "eval_loss": 1.4230467081069946, | |
| "eval_norm_levenshtein_similarity": 98.29925818708159, | |
| "eval_runtime": 350.1125, | |
| "eval_samples_per_second": 1.648, | |
| "eval_steps_per_second": 0.414, | |
| "eval_wer": 2.297165200391007, | |
| "step": 1620 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 1620, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.47782489309184e+18, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |