| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 100.0, | |
| "global_step": 21200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 1.7149999999999999e-06, | |
| "loss": 12.1703, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 3.465e-06, | |
| "loss": 9.0128, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 5.214999999999999e-06, | |
| "loss": 5.5165, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 6.964999999999999e-06, | |
| "loss": 4.6455, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 8.714999999999999e-06, | |
| "loss": 4.1839, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 1.0464999999999999e-05, | |
| "loss": 3.8664, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 1.2214999999999997e-05, | |
| "loss": 3.5691, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 1.3965e-05, | |
| "loss": 3.3224, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 1.5714999999999998e-05, | |
| "loss": 3.1431, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 1.7465e-05, | |
| "loss": 3.0387, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "eval_loss": 3.0891902446746826, | |
| "eval_runtime": 96.3286, | |
| "eval_samples_per_second": 31.538, | |
| "eval_steps_per_second": 0.986, | |
| "eval_wer": 1.0, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "learning_rate": 1.9215e-05, | |
| "loss": 2.9827, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 2.0964999999999997e-05, | |
| "loss": 2.8891, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "learning_rate": 2.2715e-05, | |
| "loss": 2.8309, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 6.6, | |
| "learning_rate": 2.4464999999999997e-05, | |
| "loss": 2.8077, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "learning_rate": 2.6215e-05, | |
| "loss": 2.7824, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.55, | |
| "learning_rate": 2.7965e-05, | |
| "loss": 2.7682, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 2.9714999999999996e-05, | |
| "loss": 2.7261, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "learning_rate": 3.1465e-05, | |
| "loss": 2.4865, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "learning_rate": 3.321499999999999e-05, | |
| "loss": 2.0721, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 9.43, | |
| "learning_rate": 3.4964999999999995e-05, | |
| "loss": 1.7911, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 9.43, | |
| "eval_loss": 0.845119059085846, | |
| "eval_runtime": 95.4637, | |
| "eval_samples_per_second": 31.824, | |
| "eval_steps_per_second": 0.995, | |
| "eval_wer": 0.6702038536721586, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 9.91, | |
| "learning_rate": 3.6714999999999997e-05, | |
| "loss": 1.667, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 10.38, | |
| "learning_rate": 3.8465e-05, | |
| "loss": 1.5783, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 10.85, | |
| "learning_rate": 4.0215e-05, | |
| "loss": 1.5165, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 11.32, | |
| "learning_rate": 4.1964999999999995e-05, | |
| "loss": 1.4551, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 11.79, | |
| "learning_rate": 4.3715e-05, | |
| "loss": 1.4246, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 12.26, | |
| "learning_rate": 4.546499999999999e-05, | |
| "loss": 1.3976, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 12.74, | |
| "learning_rate": 4.7214999999999994e-05, | |
| "loss": 1.3508, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 13.21, | |
| "learning_rate": 4.8964999999999996e-05, | |
| "loss": 1.3231, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 13.68, | |
| "learning_rate": 5.0715e-05, | |
| "loss": 1.319, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 14.15, | |
| "learning_rate": 5.2465e-05, | |
| "loss": 1.2826, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 14.15, | |
| "eval_loss": 0.4210692048072815, | |
| "eval_runtime": 96.3761, | |
| "eval_samples_per_second": 31.522, | |
| "eval_steps_per_second": 0.986, | |
| "eval_wer": 0.41658754537838594, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 14.62, | |
| "learning_rate": 5.4214999999999995e-05, | |
| "loss": 1.279, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 15.09, | |
| "learning_rate": 5.5965e-05, | |
| "loss": 1.2713, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 15.57, | |
| "learning_rate": 5.7715e-05, | |
| "loss": 1.2416, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 16.04, | |
| "learning_rate": 5.9464999999999994e-05, | |
| "loss": 1.2412, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 16.51, | |
| "learning_rate": 6.1215e-05, | |
| "loss": 1.2136, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 16.98, | |
| "learning_rate": 6.296499999999999e-05, | |
| "loss": 1.2282, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 17.45, | |
| "learning_rate": 6.471499999999999e-05, | |
| "loss": 1.1977, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 17.92, | |
| "learning_rate": 6.6465e-05, | |
| "loss": 1.204, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 18.4, | |
| "learning_rate": 6.8215e-05, | |
| "loss": 1.1807, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 18.87, | |
| "learning_rate": 6.9965e-05, | |
| "loss": 1.1802, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 18.87, | |
| "eval_loss": 0.35076606273651123, | |
| "eval_runtime": 97.3621, | |
| "eval_samples_per_second": 31.203, | |
| "eval_steps_per_second": 0.976, | |
| "eval_wer": 0.468975146607093, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 19.34, | |
| "learning_rate": 6.960116279069767e-05, | |
| "loss": 1.1843, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 19.81, | |
| "learning_rate": 6.919418604651161e-05, | |
| "loss": 1.1726, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 20.28, | |
| "learning_rate": 6.879127906976743e-05, | |
| "loss": 1.1681, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 20.75, | |
| "learning_rate": 6.838430232558139e-05, | |
| "loss": 1.1553, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 21.23, | |
| "learning_rate": 6.79813953488372e-05, | |
| "loss": 1.1477, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 21.7, | |
| "learning_rate": 6.757441860465116e-05, | |
| "loss": 1.1251, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 22.17, | |
| "learning_rate": 6.71674418604651e-05, | |
| "loss": 1.1271, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 22.64, | |
| "learning_rate": 6.676046511627906e-05, | |
| "loss": 1.1251, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 23.11, | |
| "learning_rate": 6.635755813953488e-05, | |
| "loss": 1.1315, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 23.58, | |
| "learning_rate": 6.595058139534883e-05, | |
| "loss": 1.1065, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 23.58, | |
| "eval_loss": 0.3318960666656494, | |
| "eval_runtime": 96.472, | |
| "eval_samples_per_second": 31.491, | |
| "eval_steps_per_second": 0.985, | |
| "eval_wer": 0.46618263055012565, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 24.06, | |
| "learning_rate": 6.554360465116279e-05, | |
| "loss": 1.1061, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 24.53, | |
| "learning_rate": 6.513662790697673e-05, | |
| "loss": 1.1054, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 6.472965116279069e-05, | |
| "loss": 1.0993, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 25.47, | |
| "learning_rate": 6.432267441860465e-05, | |
| "loss": 1.1, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 25.94, | |
| "learning_rate": 6.39156976744186e-05, | |
| "loss": 1.1019, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 26.42, | |
| "learning_rate": 6.350872093023256e-05, | |
| "loss": 1.078, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 26.89, | |
| "learning_rate": 6.31017441860465e-05, | |
| "loss": 1.0847, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 27.36, | |
| "learning_rate": 6.269476744186046e-05, | |
| "loss": 1.0744, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 27.83, | |
| "learning_rate": 6.228779069767442e-05, | |
| "loss": 1.0703, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 28.3, | |
| "learning_rate": 6.188081395348836e-05, | |
| "loss": 1.0921, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 28.3, | |
| "eval_loss": 0.30558520555496216, | |
| "eval_runtime": 94.8784, | |
| "eval_samples_per_second": 32.02, | |
| "eval_steps_per_second": 1.001, | |
| "eval_wer": 0.3879921809550405, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 28.77, | |
| "learning_rate": 6.147383720930232e-05, | |
| "loss": 1.0704, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 29.25, | |
| "learning_rate": 6.106686046511627e-05, | |
| "loss": 1.0479, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 29.72, | |
| "learning_rate": 6.0659883720930225e-05, | |
| "loss": 1.0583, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 30.19, | |
| "learning_rate": 6.0252906976744175e-05, | |
| "loss": 1.0515, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 30.66, | |
| "learning_rate": 5.984593023255814e-05, | |
| "loss": 1.0531, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 31.13, | |
| "learning_rate": 5.943895348837209e-05, | |
| "loss": 1.0505, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 31.6, | |
| "learning_rate": 5.9031976744186045e-05, | |
| "loss": 1.0314, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 32.08, | |
| "learning_rate": 5.8624999999999995e-05, | |
| "loss": 1.0473, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 32.55, | |
| "learning_rate": 5.821802325581395e-05, | |
| "loss": 1.0362, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 33.02, | |
| "learning_rate": 5.78110465116279e-05, | |
| "loss": 1.0366, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 33.02, | |
| "eval_loss": 0.29966893792152405, | |
| "eval_runtime": 94.2922, | |
| "eval_samples_per_second": 32.219, | |
| "eval_steps_per_second": 1.008, | |
| "eval_wer": 0.36654565763753144, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 33.49, | |
| "learning_rate": 5.740406976744185e-05, | |
| "loss": 1.0361, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 33.96, | |
| "learning_rate": 5.699709302325581e-05, | |
| "loss": 1.0298, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 34.43, | |
| "learning_rate": 5.659011627906976e-05, | |
| "loss": 1.0185, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 34.91, | |
| "learning_rate": 5.6183139534883716e-05, | |
| "loss": 1.0161, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 35.38, | |
| "learning_rate": 5.5776162790697666e-05, | |
| "loss": 1.0134, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 35.85, | |
| "learning_rate": 5.536918604651163e-05, | |
| "loss": 1.0138, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 36.32, | |
| "learning_rate": 5.496220930232558e-05, | |
| "loss": 1.0076, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 36.79, | |
| "learning_rate": 5.455523255813953e-05, | |
| "loss": 1.0134, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 37.26, | |
| "learning_rate": 5.415232558139534e-05, | |
| "loss": 0.9983, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 37.74, | |
| "learning_rate": 5.37453488372093e-05, | |
| "loss": 0.9988, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 37.74, | |
| "eval_loss": 0.2971556782722473, | |
| "eval_runtime": 94.706, | |
| "eval_samples_per_second": 32.078, | |
| "eval_steps_per_second": 1.003, | |
| "eval_wer": 0.36526110025132646, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 38.21, | |
| "learning_rate": 5.333837209302325e-05, | |
| "loss": 0.9974, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 38.68, | |
| "learning_rate": 5.293139534883721e-05, | |
| "loss": 0.9906, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 39.15, | |
| "learning_rate": 5.252441860465116e-05, | |
| "loss": 1.0178, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 39.62, | |
| "learning_rate": 5.211744186046511e-05, | |
| "loss": 0.99, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 40.09, | |
| "learning_rate": 5.171046511627907e-05, | |
| "loss": 0.9983, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 40.57, | |
| "learning_rate": 5.130348837209302e-05, | |
| "loss": 0.9835, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 41.04, | |
| "learning_rate": 5.0896511627906974e-05, | |
| "loss": 0.9902, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 41.51, | |
| "learning_rate": 5.0489534883720924e-05, | |
| "loss": 0.9604, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 41.98, | |
| "learning_rate": 5.0082558139534874e-05, | |
| "loss": 0.9836, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 42.45, | |
| "learning_rate": 4.967558139534883e-05, | |
| "loss": 0.9864, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 42.45, | |
| "eval_loss": 0.2696739435195923, | |
| "eval_runtime": 94.7296, | |
| "eval_samples_per_second": 32.07, | |
| "eval_steps_per_second": 1.003, | |
| "eval_wer": 0.33705668807595646, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 42.92, | |
| "learning_rate": 4.926860465116279e-05, | |
| "loss": 0.9692, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 43.4, | |
| "learning_rate": 4.8861627906976744e-05, | |
| "loss": 0.9643, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 43.87, | |
| "learning_rate": 4.8454651162790694e-05, | |
| "loss": 0.9682, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 44.34, | |
| "learning_rate": 4.804767441860465e-05, | |
| "loss": 0.9651, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 44.81, | |
| "learning_rate": 4.76406976744186e-05, | |
| "loss": 0.9409, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 45.28, | |
| "learning_rate": 4.723372093023255e-05, | |
| "loss": 0.9792, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 45.75, | |
| "learning_rate": 4.682674418604651e-05, | |
| "loss": 0.9405, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 46.23, | |
| "learning_rate": 4.641976744186046e-05, | |
| "loss": 0.9679, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 46.7, | |
| "learning_rate": 4.6012790697674415e-05, | |
| "loss": 0.9468, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 47.17, | |
| "learning_rate": 4.5605813953488365e-05, | |
| "loss": 0.9558, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 47.17, | |
| "eval_loss": 0.27388861775398254, | |
| "eval_runtime": 94.8478, | |
| "eval_samples_per_second": 32.03, | |
| "eval_steps_per_second": 1.002, | |
| "eval_wer": 0.314102206087685, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 47.64, | |
| "learning_rate": 4.519883720930233e-05, | |
| "loss": 0.95, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 48.11, | |
| "learning_rate": 4.479186046511628e-05, | |
| "loss": 0.9574, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 48.58, | |
| "learning_rate": 4.438895348837209e-05, | |
| "loss": 0.9473, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 49.06, | |
| "learning_rate": 4.398197674418604e-05, | |
| "loss": 0.9319, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 49.53, | |
| "learning_rate": 4.3575e-05, | |
| "loss": 0.9413, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 4.316802325581395e-05, | |
| "loss": 0.9442, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 50.47, | |
| "learning_rate": 4.27610465116279e-05, | |
| "loss": 0.9315, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 50.94, | |
| "learning_rate": 4.235406976744186e-05, | |
| "loss": 0.9334, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 51.42, | |
| "learning_rate": 4.194709302325581e-05, | |
| "loss": 0.928, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 51.89, | |
| "learning_rate": 4.1540116279069766e-05, | |
| "loss": 0.9094, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 51.89, | |
| "eval_loss": 0.2657303512096405, | |
| "eval_runtime": 93.9674, | |
| "eval_samples_per_second": 32.33, | |
| "eval_steps_per_second": 1.011, | |
| "eval_wer": 0.35330913152750626, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 52.36, | |
| "learning_rate": 4.1133139534883716e-05, | |
| "loss": 0.917, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 52.83, | |
| "learning_rate": 4.072616279069767e-05, | |
| "loss": 0.9164, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 53.3, | |
| "learning_rate": 4.031918604651162e-05, | |
| "loss": 0.915, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 53.77, | |
| "learning_rate": 3.991220930232557e-05, | |
| "loss": 0.905, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 54.25, | |
| "learning_rate": 3.950523255813953e-05, | |
| "loss": 0.9144, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 54.72, | |
| "learning_rate": 3.9098255813953487e-05, | |
| "loss": 0.8963, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 55.19, | |
| "learning_rate": 3.8691279069767443e-05, | |
| "loss": 0.8935, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 55.66, | |
| "learning_rate": 3.8284302325581393e-05, | |
| "loss": 0.8964, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 56.13, | |
| "learning_rate": 3.7877325581395344e-05, | |
| "loss": 0.9213, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 56.6, | |
| "learning_rate": 3.74703488372093e-05, | |
| "loss": 0.9034, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 56.6, | |
| "eval_loss": 0.2699211537837982, | |
| "eval_runtime": 95.9143, | |
| "eval_samples_per_second": 31.674, | |
| "eval_steps_per_second": 0.99, | |
| "eval_wer": 0.3396816531695057, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 57.08, | |
| "learning_rate": 3.706337209302325e-05, | |
| "loss": 0.8992, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 57.55, | |
| "learning_rate": 3.665639534883721e-05, | |
| "loss": 0.8911, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 58.02, | |
| "learning_rate": 3.624941860465116e-05, | |
| "loss": 0.8907, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 58.49, | |
| "learning_rate": 3.5842441860465114e-05, | |
| "loss": 0.8957, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 58.96, | |
| "learning_rate": 3.5439534883720924e-05, | |
| "loss": 0.8888, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 59.43, | |
| "learning_rate": 3.503255813953488e-05, | |
| "loss": 0.8835, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 59.91, | |
| "learning_rate": 3.462558139534883e-05, | |
| "loss": 0.8866, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 60.38, | |
| "learning_rate": 3.421860465116279e-05, | |
| "loss": 0.8887, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 60.85, | |
| "learning_rate": 3.3811627906976744e-05, | |
| "loss": 0.8919, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 61.32, | |
| "learning_rate": 3.3404651162790694e-05, | |
| "loss": 0.8907, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 61.32, | |
| "eval_loss": 0.2764558792114258, | |
| "eval_runtime": 94.1878, | |
| "eval_samples_per_second": 32.255, | |
| "eval_steps_per_second": 1.009, | |
| "eval_wer": 0.3469980452387601, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 61.79, | |
| "learning_rate": 3.2997674418604644e-05, | |
| "loss": 0.8729, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 62.26, | |
| "learning_rate": 3.25906976744186e-05, | |
| "loss": 0.8808, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 62.74, | |
| "learning_rate": 3.218372093023256e-05, | |
| "loss": 0.8845, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 63.21, | |
| "learning_rate": 3.177674418604651e-05, | |
| "loss": 0.8616, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 63.68, | |
| "learning_rate": 3.1369767441860465e-05, | |
| "loss": 0.8755, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 64.15, | |
| "learning_rate": 3.0962790697674415e-05, | |
| "loss": 0.8808, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 64.62, | |
| "learning_rate": 3.055988372093023e-05, | |
| "loss": 0.8823, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 65.09, | |
| "learning_rate": 3.0152906976744185e-05, | |
| "loss": 0.8667, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 65.57, | |
| "learning_rate": 2.974593023255814e-05, | |
| "loss": 0.8628, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 66.04, | |
| "learning_rate": 2.933895348837209e-05, | |
| "loss": 0.8631, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 66.04, | |
| "eval_loss": 0.27741023898124695, | |
| "eval_runtime": 93.7633, | |
| "eval_samples_per_second": 32.401, | |
| "eval_steps_per_second": 1.013, | |
| "eval_wer": 0.33459927394582517, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 66.51, | |
| "learning_rate": 2.8931976744186042e-05, | |
| "loss": 0.865, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 66.98, | |
| "learning_rate": 2.8524999999999995e-05, | |
| "loss": 0.857, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 67.45, | |
| "learning_rate": 2.8118023255813952e-05, | |
| "loss": 0.8575, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 67.92, | |
| "learning_rate": 2.7711046511627906e-05, | |
| "loss": 0.8497, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 68.4, | |
| "learning_rate": 2.730406976744186e-05, | |
| "loss": 0.855, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 68.87, | |
| "learning_rate": 2.689709302325581e-05, | |
| "loss": 0.845, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 69.34, | |
| "learning_rate": 2.6490116279069763e-05, | |
| "loss": 0.8324, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 69.81, | |
| "learning_rate": 2.608313953488372e-05, | |
| "loss": 0.844, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 70.28, | |
| "learning_rate": 2.5676162790697673e-05, | |
| "loss": 0.8483, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 70.75, | |
| "learning_rate": 2.5269186046511626e-05, | |
| "loss": 0.8389, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 70.75, | |
| "eval_loss": 0.2742805778980255, | |
| "eval_runtime": 94.8277, | |
| "eval_samples_per_second": 32.037, | |
| "eval_steps_per_second": 1.002, | |
| "eval_wer": 0.33649818486456295, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 71.23, | |
| "learning_rate": 2.486220930232558e-05, | |
| "loss": 0.8383, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 71.7, | |
| "learning_rate": 2.445523255813953e-05, | |
| "loss": 0.8468, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 72.17, | |
| "learning_rate": 2.4048255813953487e-05, | |
| "loss": 0.8511, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 72.64, | |
| "learning_rate": 2.364127906976744e-05, | |
| "loss": 0.8457, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 73.11, | |
| "learning_rate": 2.3234302325581394e-05, | |
| "loss": 0.8314, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 73.58, | |
| "learning_rate": 2.2827325581395347e-05, | |
| "loss": 0.8278, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 74.06, | |
| "learning_rate": 2.24203488372093e-05, | |
| "loss": 0.8481, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 74.53, | |
| "learning_rate": 2.201337209302325e-05, | |
| "loss": 0.8158, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "learning_rate": 2.1606395348837207e-05, | |
| "loss": 0.8356, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 75.47, | |
| "learning_rate": 2.119941860465116e-05, | |
| "loss": 0.8214, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 75.47, | |
| "eval_loss": 0.2778330445289612, | |
| "eval_runtime": 94.0407, | |
| "eval_samples_per_second": 32.305, | |
| "eval_steps_per_second": 1.01, | |
| "eval_wer": 0.32013404077073443, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 75.94, | |
| "learning_rate": 2.0792441860465114e-05, | |
| "loss": 0.8293, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 76.42, | |
| "learning_rate": 2.0389534883720927e-05, | |
| "loss": 0.834, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 76.89, | |
| "learning_rate": 1.9982558139534884e-05, | |
| "loss": 0.8308, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 77.36, | |
| "learning_rate": 1.9575581395348838e-05, | |
| "loss": 0.8292, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 77.83, | |
| "learning_rate": 1.9168604651162788e-05, | |
| "loss": 0.8224, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 78.3, | |
| "learning_rate": 1.876162790697674e-05, | |
| "loss": 0.8172, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 78.77, | |
| "learning_rate": 1.8354651162790695e-05, | |
| "loss": 0.8202, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 79.25, | |
| "learning_rate": 1.795174418604651e-05, | |
| "loss": 0.8222, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 79.72, | |
| "learning_rate": 1.7544767441860464e-05, | |
| "loss": 0.811, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 80.19, | |
| "learning_rate": 1.7137790697674414e-05, | |
| "loss": 0.8195, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 80.19, | |
| "eval_loss": 0.2725354731082916, | |
| "eval_runtime": 94.0258, | |
| "eval_samples_per_second": 32.31, | |
| "eval_steps_per_second": 1.01, | |
| "eval_wer": 0.32856743926277576, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 80.66, | |
| "learning_rate": 1.673081395348837e-05, | |
| "loss": 0.8053, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 81.13, | |
| "learning_rate": 1.6323837209302325e-05, | |
| "loss": 0.8152, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 81.6, | |
| "learning_rate": 1.5916860465116278e-05, | |
| "loss": 0.8074, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 82.08, | |
| "learning_rate": 1.550988372093023e-05, | |
| "loss": 0.8179, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 82.55, | |
| "learning_rate": 1.5102906976744183e-05, | |
| "loss": 0.8078, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 83.02, | |
| "learning_rate": 1.4695930232558139e-05, | |
| "loss": 0.8078, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 83.49, | |
| "learning_rate": 1.4288953488372092e-05, | |
| "loss": 0.8061, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 83.96, | |
| "learning_rate": 1.3881976744186044e-05, | |
| "loss": 0.813, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 84.43, | |
| "learning_rate": 1.3474999999999999e-05, | |
| "loss": 0.7891, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 84.91, | |
| "learning_rate": 1.3068023255813952e-05, | |
| "loss": 0.7994, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 84.91, | |
| "eval_loss": 0.2781510651111603, | |
| "eval_runtime": 92.9473, | |
| "eval_samples_per_second": 32.685, | |
| "eval_steps_per_second": 1.022, | |
| "eval_wer": 0.33147165596202177, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 85.38, | |
| "learning_rate": 1.2661046511627907e-05, | |
| "loss": 0.8011, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 85.85, | |
| "learning_rate": 1.225406976744186e-05, | |
| "loss": 0.7886, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 86.32, | |
| "learning_rate": 1.1847093023255813e-05, | |
| "loss": 0.8107, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 86.79, | |
| "learning_rate": 1.1440116279069768e-05, | |
| "loss": 0.7976, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 87.26, | |
| "learning_rate": 1.103313953488372e-05, | |
| "loss": 0.7871, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 87.74, | |
| "learning_rate": 1.0626162790697673e-05, | |
| "loss": 0.7922, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 88.21, | |
| "learning_rate": 1.0219186046511628e-05, | |
| "loss": 0.7826, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 88.68, | |
| "learning_rate": 9.81220930232558e-06, | |
| "loss": 0.7901, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 89.15, | |
| "learning_rate": 9.405232558139535e-06, | |
| "loss": 0.793, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 89.62, | |
| "learning_rate": 8.998255813953488e-06, | |
| "loss": 0.7816, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 89.62, | |
| "eval_loss": 0.2774805724620819, | |
| "eval_runtime": 94.1962, | |
| "eval_samples_per_second": 32.252, | |
| "eval_steps_per_second": 1.009, | |
| "eval_wer": 0.3363306339011449, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 90.09, | |
| "learning_rate": 8.591279069767442e-06, | |
| "loss": 0.7922, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 90.57, | |
| "learning_rate": 8.188372093023255e-06, | |
| "loss": 0.7937, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 91.04, | |
| "learning_rate": 7.781395348837208e-06, | |
| "loss": 0.784, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 91.51, | |
| "learning_rate": 7.374418604651162e-06, | |
| "loss": 0.7811, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 91.98, | |
| "learning_rate": 6.967441860465116e-06, | |
| "loss": 0.7784, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 92.45, | |
| "learning_rate": 6.560465116279069e-06, | |
| "loss": 0.789, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 92.92, | |
| "learning_rate": 6.153488372093022e-06, | |
| "loss": 0.7785, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 93.4, | |
| "learning_rate": 5.7465116279069765e-06, | |
| "loss": 0.7799, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 93.87, | |
| "learning_rate": 5.33953488372093e-06, | |
| "loss": 0.7784, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 94.34, | |
| "learning_rate": 4.9325581395348825e-06, | |
| "loss": 0.7816, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 94.34, | |
| "eval_loss": 0.2731296420097351, | |
| "eval_runtime": 94.2664, | |
| "eval_samples_per_second": 32.228, | |
| "eval_steps_per_second": 1.008, | |
| "eval_wer": 0.32784138508796423, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 94.81, | |
| "learning_rate": 4.525581395348837e-06, | |
| "loss": 0.7788, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 95.28, | |
| "learning_rate": 4.11860465116279e-06, | |
| "loss": 0.781, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 95.75, | |
| "learning_rate": 3.7116279069767437e-06, | |
| "loss": 0.785, | |
| "step": 20300 | |
| }, | |
| { | |
| "epoch": 96.23, | |
| "learning_rate": 3.3046511627906976e-06, | |
| "loss": 0.7664, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 96.7, | |
| "learning_rate": 2.8976744186046506e-06, | |
| "loss": 0.7771, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 97.17, | |
| "learning_rate": 2.4906976744186045e-06, | |
| "loss": 0.7848, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 97.64, | |
| "learning_rate": 2.083720930232558e-06, | |
| "loss": 0.765, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 98.11, | |
| "learning_rate": 1.6767441860465114e-06, | |
| "loss": 0.7877, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 98.58, | |
| "learning_rate": 1.2697674418604648e-06, | |
| "loss": 0.7681, | |
| "step": 20900 | |
| }, | |
| { | |
| "epoch": 99.06, | |
| "learning_rate": 8.627906976744186e-07, | |
| "loss": 0.7635, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 99.06, | |
| "eval_loss": 0.276745080947876, | |
| "eval_runtime": 95.4293, | |
| "eval_samples_per_second": 31.835, | |
| "eval_steps_per_second": 0.996, | |
| "eval_wer": 0.32594247416922645, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 99.53, | |
| "learning_rate": 4.558139534883721e-07, | |
| "loss": 0.7818, | |
| "step": 21100 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "learning_rate": 4.883720930232558e-08, | |
| "loss": 0.7716, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 21200, | |
| "total_flos": 7.799992701307658e+19, | |
| "train_loss": 1.235663890838623, | |
| "train_runtime": 30132.3883, | |
| "train_samples_per_second": 22.464, | |
| "train_steps_per_second": 0.704 | |
| } | |
| ], | |
| "max_steps": 21200, | |
| "num_train_epochs": 100, | |
| "total_flos": 7.799992701307658e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |