| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 19.99970457902511, | |
| "global_step": 16920, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00029775, | |
| "loss": 3.3587, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "eval_loss": 1.1883399486541748, | |
| "eval_runtime": 298.8414, | |
| "eval_samples_per_second": 16.986, | |
| "eval_steps_per_second": 2.831, | |
| "eval_wer": 0.8392195865162833, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.00029279055690072635, | |
| "loss": 1.8377, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 0.8830727338790894, | |
| "eval_runtime": 295.1554, | |
| "eval_samples_per_second": 17.198, | |
| "eval_steps_per_second": 2.866, | |
| "eval_wer": 0.6852318175085172, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 0.00028552663438256653, | |
| "loss": 1.7118, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "eval_loss": 0.8031275868415833, | |
| "eval_runtime": 296.1538, | |
| "eval_samples_per_second": 17.14, | |
| "eval_steps_per_second": 2.857, | |
| "eval_wer": 0.6565588166832427, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.00027826271186440676, | |
| "loss": 1.6741, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_loss": 0.7517648339271545, | |
| "eval_runtime": 296.1537, | |
| "eval_samples_per_second": 17.14, | |
| "eval_steps_per_second": 2.857, | |
| "eval_wer": 0.6104492456144063, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 0.00027099878934624694, | |
| "loss": 1.6163, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "eval_loss": 0.6887586712837219, | |
| "eval_runtime": 298.4301, | |
| "eval_samples_per_second": 17.009, | |
| "eval_steps_per_second": 2.835, | |
| "eval_wer": 0.5590706138773092, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 0.0002637348668280871, | |
| "loss": 1.5782, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "eval_loss": 0.6580386161804199, | |
| "eval_runtime": 296.628, | |
| "eval_samples_per_second": 17.112, | |
| "eval_steps_per_second": 2.852, | |
| "eval_wer": 0.5164525890344288, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 0.0002564709443099273, | |
| "loss": 1.5548, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "eval_loss": 0.6505530476570129, | |
| "eval_runtime": 300.5251, | |
| "eval_samples_per_second": 16.89, | |
| "eval_steps_per_second": 2.815, | |
| "eval_wer": 0.5184417123389128, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 0.00024920702179176754, | |
| "loss": 1.5249, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "eval_loss": 0.6197642683982849, | |
| "eval_runtime": 298.2224, | |
| "eval_samples_per_second": 17.021, | |
| "eval_steps_per_second": 2.837, | |
| "eval_wer": 0.5028461391963095, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 0.00024194309927360772, | |
| "loss": 1.5078, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "eval_loss": 0.5992246866226196, | |
| "eval_runtime": 298.0744, | |
| "eval_samples_per_second": 17.029, | |
| "eval_steps_per_second": 2.838, | |
| "eval_wer": 0.4932179359671583, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 0.0002346791767554479, | |
| "loss": 1.4836, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "eval_loss": 0.5705241560935974, | |
| "eval_runtime": 292.5306, | |
| "eval_samples_per_second": 17.352, | |
| "eval_steps_per_second": 2.892, | |
| "eval_wer": 0.46513743995598533, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.2, | |
| "learning_rate": 0.0002274152542372881, | |
| "loss": 1.4505, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 5.2, | |
| "eval_loss": 0.5488837957382202, | |
| "eval_runtime": 300.7764, | |
| "eval_samples_per_second": 16.876, | |
| "eval_steps_per_second": 2.813, | |
| "eval_wer": 0.4507691982140212, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "learning_rate": 0.00022015133171912832, | |
| "loss": 1.4481, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "eval_loss": 0.5577110052108765, | |
| "eval_runtime": 299.1357, | |
| "eval_samples_per_second": 16.969, | |
| "eval_steps_per_second": 2.828, | |
| "eval_wer": 0.45624986774446114, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 0.0002128874092009685, | |
| "loss": 1.4136, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "eval_loss": 0.5452219247817993, | |
| "eval_runtime": 298.6905, | |
| "eval_samples_per_second": 16.994, | |
| "eval_steps_per_second": 2.832, | |
| "eval_wer": 0.4370992657172482, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 0.00020564164648910412, | |
| "loss": 1.3861, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "eval_loss": 0.510087788105011, | |
| "eval_runtime": 298.7447, | |
| "eval_samples_per_second": 16.991, | |
| "eval_steps_per_second": 2.832, | |
| "eval_wer": 0.40865903464037073, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "learning_rate": 0.0001983777239709443, | |
| "loss": 1.3772, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "eval_loss": 0.49333110451698303, | |
| "eval_runtime": 298.2699, | |
| "eval_samples_per_second": 17.018, | |
| "eval_steps_per_second": 2.836, | |
| "eval_wer": 0.39513722834712317, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.56, | |
| "learning_rate": 0.00019111380145278448, | |
| "loss": 1.3478, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 7.56, | |
| "eval_loss": 0.4849308133125305, | |
| "eval_runtime": 299.279, | |
| "eval_samples_per_second": 16.961, | |
| "eval_steps_per_second": 2.827, | |
| "eval_wer": 0.3921747042766151, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "learning_rate": 0.0001838498789346247, | |
| "loss": 1.3394, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "eval_loss": 0.4805210828781128, | |
| "eval_runtime": 297.1556, | |
| "eval_samples_per_second": 17.082, | |
| "eval_steps_per_second": 2.847, | |
| "eval_wer": 0.3891698584336712, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 8.51, | |
| "learning_rate": 0.00017658595641646487, | |
| "loss": 1.3095, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 8.51, | |
| "eval_loss": 0.48388615250587463, | |
| "eval_runtime": 298.6068, | |
| "eval_samples_per_second": 16.999, | |
| "eval_steps_per_second": 2.833, | |
| "eval_wer": 0.38335061472374465, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "learning_rate": 0.00016932203389830508, | |
| "loss": 1.306, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "eval_loss": 0.461063027381897, | |
| "eval_runtime": 296.7161, | |
| "eval_samples_per_second": 17.107, | |
| "eval_steps_per_second": 2.851, | |
| "eval_wer": 0.35865586050743803, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 9.46, | |
| "learning_rate": 0.00016205811138014526, | |
| "loss": 1.2707, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 9.46, | |
| "eval_loss": 0.4544948935508728, | |
| "eval_runtime": 298.2639, | |
| "eval_samples_per_second": 17.018, | |
| "eval_steps_per_second": 2.836, | |
| "eval_wer": 0.3730241022494022, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 9.93, | |
| "learning_rate": 0.00015479418886198547, | |
| "loss": 1.2626, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 9.93, | |
| "eval_loss": 0.4515869617462158, | |
| "eval_runtime": 299.4234, | |
| "eval_samples_per_second": 16.953, | |
| "eval_steps_per_second": 2.825, | |
| "eval_wer": 0.35241339907315317, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 10.4, | |
| "learning_rate": 0.00014753026634382565, | |
| "loss": 1.2412, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 10.4, | |
| "eval_loss": 0.4314204454421997, | |
| "eval_runtime": 294.8808, | |
| "eval_samples_per_second": 17.214, | |
| "eval_steps_per_second": 2.869, | |
| "eval_wer": 0.3310197431068413, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 10.87, | |
| "learning_rate": 0.00014026634382566583, | |
| "loss": 1.2456, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 10.87, | |
| "eval_loss": 0.4400792419910431, | |
| "eval_runtime": 296.6883, | |
| "eval_samples_per_second": 17.109, | |
| "eval_steps_per_second": 2.851, | |
| "eval_wer": 0.3458958461180354, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 11.35, | |
| "learning_rate": 0.00013302058111380144, | |
| "loss": 1.2081, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 11.35, | |
| "eval_loss": 0.4399240016937256, | |
| "eval_runtime": 295.1566, | |
| "eval_samples_per_second": 17.198, | |
| "eval_steps_per_second": 2.866, | |
| "eval_wer": 0.33563281630234676, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 11.82, | |
| "learning_rate": 0.00012575665859564165, | |
| "loss": 1.1998, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 11.82, | |
| "eval_loss": 0.41947221755981445, | |
| "eval_runtime": 296.8576, | |
| "eval_samples_per_second": 17.099, | |
| "eval_steps_per_second": 2.85, | |
| "eval_wer": 0.3215396660812155, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 12.29, | |
| "learning_rate": 0.00011851089588377723, | |
| "loss": 1.1826, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 12.29, | |
| "eval_loss": 0.4220864474773407, | |
| "eval_runtime": 296.3734, | |
| "eval_samples_per_second": 17.127, | |
| "eval_steps_per_second": 2.855, | |
| "eval_wer": 0.3177518674482087, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 12.77, | |
| "learning_rate": 0.00011124697336561742, | |
| "loss": 1.1573, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 12.77, | |
| "eval_loss": 0.4097737967967987, | |
| "eval_runtime": 297.6541, | |
| "eval_samples_per_second": 17.053, | |
| "eval_steps_per_second": 2.842, | |
| "eval_wer": 0.3084199166261083, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 13.24, | |
| "learning_rate": 0.00010398305084745762, | |
| "loss": 1.1416, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 13.24, | |
| "eval_loss": 0.4085627496242523, | |
| "eval_runtime": 296.7019, | |
| "eval_samples_per_second": 17.108, | |
| "eval_steps_per_second": 2.851, | |
| "eval_wer": 0.3119114628520642, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 13.71, | |
| "learning_rate": 9.671912832929781e-05, | |
| "loss": 1.1174, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 13.71, | |
| "eval_loss": 0.3854358196258545, | |
| "eval_runtime": 297.2728, | |
| "eval_samples_per_second": 17.075, | |
| "eval_steps_per_second": 2.846, | |
| "eval_wer": 0.2910468290412003, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 14.18, | |
| "learning_rate": 8.947336561743341e-05, | |
| "loss": 1.1048, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 14.18, | |
| "eval_loss": 0.38590070605278015, | |
| "eval_runtime": 297.414, | |
| "eval_samples_per_second": 17.067, | |
| "eval_steps_per_second": 2.845, | |
| "eval_wer": 0.282413187464291, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 14.66, | |
| "learning_rate": 8.22094430992736e-05, | |
| "loss": 1.0748, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 14.66, | |
| "eval_loss": 0.3854171633720398, | |
| "eval_runtime": 295.6597, | |
| "eval_samples_per_second": 17.168, | |
| "eval_steps_per_second": 2.861, | |
| "eval_wer": 0.275705186533212, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 15.13, | |
| "learning_rate": 7.494552058111379e-05, | |
| "loss": 1.0697, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 15.13, | |
| "eval_loss": 0.37401217222213745, | |
| "eval_runtime": 297.4379, | |
| "eval_samples_per_second": 17.066, | |
| "eval_steps_per_second": 2.844, | |
| "eval_wer": 0.2723829273969994, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 15.6, | |
| "learning_rate": 6.7681598062954e-05, | |
| "loss": 1.0477, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 15.6, | |
| "eval_loss": 0.3693487048149109, | |
| "eval_runtime": 298.4761, | |
| "eval_samples_per_second": 17.006, | |
| "eval_steps_per_second": 2.834, | |
| "eval_wer": 0.2642994688617559, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 16.08, | |
| "learning_rate": 6.0417675544794184e-05, | |
| "loss": 1.0356, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 16.08, | |
| "eval_loss": 0.3726678788661957, | |
| "eval_runtime": 298.583, | |
| "eval_samples_per_second": 17.0, | |
| "eval_steps_per_second": 2.833, | |
| "eval_wer": 0.2561102058954229, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 16.55, | |
| "learning_rate": 5.315375302663438e-05, | |
| "loss": 1.0083, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 16.55, | |
| "eval_loss": 0.36522331833839417, | |
| "eval_runtime": 301.7885, | |
| "eval_samples_per_second": 16.82, | |
| "eval_steps_per_second": 2.803, | |
| "eval_wer": 0.25007935332331716, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 17.02, | |
| "learning_rate": 4.5889830508474574e-05, | |
| "loss": 1.0, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 17.02, | |
| "eval_loss": 0.36411818861961365, | |
| "eval_runtime": 297.5091, | |
| "eval_samples_per_second": 17.062, | |
| "eval_steps_per_second": 2.844, | |
| "eval_wer": 0.24572021076242673, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 17.49, | |
| "learning_rate": 3.862590799031477e-05, | |
| "loss": 0.9779, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 17.49, | |
| "eval_loss": 0.35678642988204956, | |
| "eval_runtime": 295.5669, | |
| "eval_samples_per_second": 17.174, | |
| "eval_steps_per_second": 2.862, | |
| "eval_wer": 0.24089552870474215, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 17.97, | |
| "learning_rate": 3.1361985472154963e-05, | |
| "loss": 0.9596, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 17.97, | |
| "eval_loss": 0.3557918071746826, | |
| "eval_runtime": 296.6965, | |
| "eval_samples_per_second": 17.108, | |
| "eval_steps_per_second": 2.851, | |
| "eval_wer": 0.23759443045474746, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 18.44, | |
| "learning_rate": 2.4098062953995155e-05, | |
| "loss": 0.946, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 18.44, | |
| "eval_loss": 0.35907429456710815, | |
| "eval_runtime": 295.7242, | |
| "eval_samples_per_second": 17.165, | |
| "eval_steps_per_second": 2.861, | |
| "eval_wer": 0.23114036015828343, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 18.91, | |
| "learning_rate": 1.685230024213075e-05, | |
| "loss": 0.9389, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 18.91, | |
| "eval_loss": 0.3540380001068115, | |
| "eval_runtime": 296.3149, | |
| "eval_samples_per_second": 17.13, | |
| "eval_steps_per_second": 2.855, | |
| "eval_wer": 0.22828364051886493, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 19.39, | |
| "learning_rate": 9.588377723970942e-06, | |
| "loss": 0.9173, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 19.39, | |
| "eval_loss": 0.35515815019607544, | |
| "eval_runtime": 296.2267, | |
| "eval_samples_per_second": 17.136, | |
| "eval_steps_per_second": 2.856, | |
| "eval_wer": 0.2265061260765601, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 19.86, | |
| "learning_rate": 2.3244552058111378e-06, | |
| "loss": 0.9122, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 19.86, | |
| "eval_loss": 0.353456050157547, | |
| "eval_runtime": 298.3571, | |
| "eval_samples_per_second": 17.013, | |
| "eval_steps_per_second": 2.836, | |
| "eval_wer": 0.22502486404130606, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 16920, | |
| "total_flos": 1.9284410553255697e+20, | |
| "train_loss": 1.3125287554224614, | |
| "train_runtime": 67690.7896, | |
| "train_samples_per_second": 6.0, | |
| "train_steps_per_second": 0.25 | |
| } | |
| ], | |
| "max_steps": 16920, | |
| "num_train_epochs": 20, | |
| "total_flos": 1.9284410553255697e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |