| { | |
| "best_metric": 5.243532975996271, | |
| "best_model_checkpoint": "./final-whisper-for-initial-publish/checkpoint-1000", | |
| "epoch": 1.6474464579901154, | |
| "eval_steps": 500, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04118616144975288, | |
| "grad_norm": 67.08393096923828, | |
| "learning_rate": 4.2000000000000006e-07, | |
| "loss": 2.8367, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.08237232289950576, | |
| "grad_norm": 17.447174072265625, | |
| "learning_rate": 9.200000000000001e-07, | |
| "loss": 2.0282, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.12355848434925865, | |
| "grad_norm": 12.4807710647583, | |
| "learning_rate": 1.42e-06, | |
| "loss": 1.3152, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.16474464579901152, | |
| "grad_norm": 8.900789260864258, | |
| "learning_rate": 1.9200000000000003e-06, | |
| "loss": 0.7248, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.20593080724876442, | |
| "grad_norm": 7.897016525268555, | |
| "learning_rate": 2.42e-06, | |
| "loss": 0.5731, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.2471169686985173, | |
| "grad_norm": 9.260579109191895, | |
| "learning_rate": 2.92e-06, | |
| "loss": 0.474, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.2883031301482702, | |
| "grad_norm": 6.941812515258789, | |
| "learning_rate": 3.4200000000000007e-06, | |
| "loss": 0.4304, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.32948929159802304, | |
| "grad_norm": 6.919547080993652, | |
| "learning_rate": 3.920000000000001e-06, | |
| "loss": 0.3514, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.37067545304777594, | |
| "grad_norm": 6.444908618927002, | |
| "learning_rate": 4.42e-06, | |
| "loss": 0.273, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.41186161449752884, | |
| "grad_norm": 5.828249454498291, | |
| "learning_rate": 4.92e-06, | |
| "loss": 0.1655, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.45304777594728174, | |
| "grad_norm": 2.492833137512207, | |
| "learning_rate": 5.420000000000001e-06, | |
| "loss": 0.0894, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.4942339373970346, | |
| "grad_norm": 3.24698805809021, | |
| "learning_rate": 5.92e-06, | |
| "loss": 0.076, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5354200988467874, | |
| "grad_norm": 4.378988265991211, | |
| "learning_rate": 6.42e-06, | |
| "loss": 0.0763, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.5766062602965404, | |
| "grad_norm": 3.146955728530884, | |
| "learning_rate": 6.92e-06, | |
| "loss": 0.0679, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.6177924217462932, | |
| "grad_norm": 3.8782553672790527, | |
| "learning_rate": 7.420000000000001e-06, | |
| "loss": 0.0524, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.6589785831960461, | |
| "grad_norm": 6.15730094909668, | |
| "learning_rate": 7.92e-06, | |
| "loss": 0.0554, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.700164744645799, | |
| "grad_norm": 6.1213812828063965, | |
| "learning_rate": 8.42e-06, | |
| "loss": 0.0535, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.7413509060955519, | |
| "grad_norm": 4.010768890380859, | |
| "learning_rate": 8.920000000000001e-06, | |
| "loss": 0.0465, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.7825370675453048, | |
| "grad_norm": 3.509321451187134, | |
| "learning_rate": 9.42e-06, | |
| "loss": 0.0436, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.8237232289950577, | |
| "grad_norm": 2.744082450866699, | |
| "learning_rate": 9.920000000000002e-06, | |
| "loss": 0.0468, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8237232289950577, | |
| "eval_loss": 0.04186907038092613, | |
| "eval_runtime": 1739.4189, | |
| "eval_samples_per_second": 1.395, | |
| "eval_steps_per_second": 0.175, | |
| "eval_wer": 10.859939408063388, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8649093904448105, | |
| "grad_norm": 2.581739664077759, | |
| "learning_rate": 9.953333333333333e-06, | |
| "loss": 0.0387, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.9060955518945635, | |
| "grad_norm": 1.7439745664596558, | |
| "learning_rate": 9.89777777777778e-06, | |
| "loss": 0.0372, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.9472817133443163, | |
| "grad_norm": 3.0741066932678223, | |
| "learning_rate": 9.842222222222223e-06, | |
| "loss": 0.0344, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.9884678747940692, | |
| "grad_norm": 2.640840530395508, | |
| "learning_rate": 9.786666666666667e-06, | |
| "loss": 0.0331, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.029654036243822, | |
| "grad_norm": 0.908171534538269, | |
| "learning_rate": 9.731111111111113e-06, | |
| "loss": 0.0267, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 1.0708401976935749, | |
| "grad_norm": 1.0449379682540894, | |
| "learning_rate": 9.675555555555555e-06, | |
| "loss": 0.0203, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.112026359143328, | |
| "grad_norm": 2.090158700942993, | |
| "learning_rate": 9.620000000000001e-06, | |
| "loss": 0.0185, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 1.1532125205930808, | |
| "grad_norm": 3.4184322357177734, | |
| "learning_rate": 9.564444444444445e-06, | |
| "loss": 0.0226, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.1943986820428336, | |
| "grad_norm": 2.6463236808776855, | |
| "learning_rate": 9.508888888888889e-06, | |
| "loss": 0.0155, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 1.2355848434925865, | |
| "grad_norm": 0.7438070774078369, | |
| "learning_rate": 9.453333333333335e-06, | |
| "loss": 0.0184, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.2767710049423393, | |
| "grad_norm": 1.7319138050079346, | |
| "learning_rate": 9.397777777777779e-06, | |
| "loss": 0.0161, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 1.3179571663920924, | |
| "grad_norm": 1.9704327583312988, | |
| "learning_rate": 9.342222222222223e-06, | |
| "loss": 0.0137, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.3591433278418452, | |
| "grad_norm": 1.808190107345581, | |
| "learning_rate": 9.286666666666667e-06, | |
| "loss": 0.0169, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 1.400329489291598, | |
| "grad_norm": 2.1106882095336914, | |
| "learning_rate": 9.231111111111111e-06, | |
| "loss": 0.0141, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.441515650741351, | |
| "grad_norm": 1.5317939519882202, | |
| "learning_rate": 9.175555555555557e-06, | |
| "loss": 0.0114, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 1.4827018121911038, | |
| "grad_norm": 2.758979082107544, | |
| "learning_rate": 9.12e-06, | |
| "loss": 0.0123, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.5238879736408566, | |
| "grad_norm": 2.9624338150024414, | |
| "learning_rate": 9.064444444444447e-06, | |
| "loss": 0.014, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 1.5650741350906094, | |
| "grad_norm": 2.520521402359009, | |
| "learning_rate": 9.008888888888889e-06, | |
| "loss": 0.0107, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.6062602965403623, | |
| "grad_norm": 1.0015959739685059, | |
| "learning_rate": 8.953333333333335e-06, | |
| "loss": 0.0111, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 1.6474464579901154, | |
| "grad_norm": 2.2675178050994873, | |
| "learning_rate": 8.897777777777779e-06, | |
| "loss": 0.0124, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.6474464579901154, | |
| "eval_loss": 0.02125450409948826, | |
| "eval_runtime": 1730.2385, | |
| "eval_samples_per_second": 1.403, | |
| "eval_steps_per_second": 0.176, | |
| "eval_wer": 5.243532975996271, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 500, | |
| "total_flos": 4.61621206351872e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |