| { | |
| "best_metric": 0.3557792992347966, | |
| "best_model_checkpoint": "wav2vec2-xlsr53-TH-cmv-ckp-farfield-40epochs/checkpoint-33000", | |
| "epoch": 40.0, | |
| "global_step": 34000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 2.8528887336629845e-06, | |
| "loss": 37.8606, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 5.73518869138435e-06, | |
| "loss": 36.6664, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 8.646899873164096e-06, | |
| "loss": 24.0296, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 1.1588022279002226e-05, | |
| "loss": 11.6622, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 1.4529144684840354e-05, | |
| "loss": 7.3831, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.747026709067848e-05, | |
| "loss": 5.378, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 2.0411389496516608e-05, | |
| "loss": 4.2757, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 2.3352511902354736e-05, | |
| "loss": 3.8626, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 2.6293634308192865e-05, | |
| "loss": 3.7576, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 2.9234756714030993e-05, | |
| "loss": 3.694, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "eval_loss": 3.646904945373535, | |
| "eval_runtime": 108.0062, | |
| "eval_samples_per_second": 39.007, | |
| "eval_steps_per_second": 0.611, | |
| "eval_wer": 1.0, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 3.217587911986912e-05, | |
| "loss": 3.6476, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 3.511700152570725e-05, | |
| "loss": 3.6198, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 3.805812393154537e-05, | |
| "loss": 3.6033, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 4.099924633738351e-05, | |
| "loss": 3.5974, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 4.3940368743221636e-05, | |
| "loss": 3.5822, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 4.6881491149059764e-05, | |
| "loss": 3.5273, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.982261355489789e-05, | |
| "loss": 3.3921, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 5.2763735960736014e-05, | |
| "loss": 3.0588, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 5.570485836657414e-05, | |
| "loss": 2.592, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 5.864598077241228e-05, | |
| "loss": 2.2453, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "eval_loss": 1.3178459405899048, | |
| "eval_runtime": 106.709, | |
| "eval_samples_per_second": 39.481, | |
| "eval_steps_per_second": 0.619, | |
| "eval_wer": 0.8124043495771245, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 6.15871031782504e-05, | |
| "loss": 2.0098, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 6.452822558408853e-05, | |
| "loss": 1.8594, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 6.746934798992666e-05, | |
| "loss": 1.7491, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 7.041047039576478e-05, | |
| "loss": 1.6576, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 7.335159280160291e-05, | |
| "loss": 1.6024, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 7.629271520744105e-05, | |
| "loss": 1.5301, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 7.923383761327917e-05, | |
| "loss": 1.4785, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "learning_rate": 8.217496001911729e-05, | |
| "loss": 1.4362, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 8.511608242495543e-05, | |
| "loss": 1.3932, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 8.805720483079356e-05, | |
| "loss": 1.3807, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "eval_loss": 0.6274150609970093, | |
| "eval_runtime": 114.1009, | |
| "eval_samples_per_second": 36.923, | |
| "eval_steps_per_second": 0.578, | |
| "eval_wer": 0.5551751913008458, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "learning_rate": 9.099832723663168e-05, | |
| "loss": 1.3404, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 9.39394496424698e-05, | |
| "loss": 1.314, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "learning_rate": 9.688057204830794e-05, | |
| "loss": 1.2968, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 9.982169445414606e-05, | |
| "loss": 1.2514, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 9.985126147657674e-05, | |
| "loss": 1.2271, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "learning_rate": 9.936705484698494e-05, | |
| "loss": 1.1943, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 9.855002472512436e-05, | |
| "loss": 1.1604, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 9.740567937595607e-05, | |
| "loss": 1.1508, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 9.59417337627618e-05, | |
| "loss": 1.1627, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "learning_rate": 9.41680575342979e-05, | |
| "loss": 1.1464, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "eval_loss": 0.5025569200515747, | |
| "eval_runtime": 105.3594, | |
| "eval_samples_per_second": 39.987, | |
| "eval_steps_per_second": 0.626, | |
| "eval_wer": 0.4815142972211035, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "learning_rate": 9.209660848546103e-05, | |
| "loss": 1.1392, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "learning_rate": 8.97413519400613e-05, | |
| "loss": 1.1173, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 8.711816659920938e-05, | |
| "loss": 1.1, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 5.18, | |
| "learning_rate": 8.424473749006998e-05, | |
| "loss": 1.0627, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 5.29, | |
| "learning_rate": 8.114043673670144e-05, | |
| "loss": 1.0712, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 5.41, | |
| "learning_rate": 7.782619295680218e-05, | |
| "loss": 1.0416, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "learning_rate": 7.432435016486675e-05, | |
| "loss": 1.0254, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "learning_rate": 7.06585171330004e-05, | |
| "loss": 1.0391, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 6.68534082249731e-05, | |
| "loss": 1.0147, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "learning_rate": 6.29346767765805e-05, | |
| "loss": 1.002, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "eval_loss": 0.44318458437919617, | |
| "eval_runtime": 108.9364, | |
| "eval_samples_per_second": 38.674, | |
| "eval_steps_per_second": 0.606, | |
| "eval_wer": 0.45110753121224323, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 5.892874214563035e-05, | |
| "loss": 0.984, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 5.4862611597551125e-05, | |
| "loss": 0.9435, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "learning_rate": 5.076369822743683e-05, | |
| "loss": 0.9651, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 6.35, | |
| "learning_rate": 4.665963614606392e-05, | |
| "loss": 0.9421, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 6.47, | |
| "learning_rate": 4.257809417586155e-05, | |
| "loss": 0.9307, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 6.59, | |
| "learning_rate": 3.854658931286227e-05, | |
| "loss": 0.925, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 6.71, | |
| "learning_rate": 3.459230121223818e-05, | |
| "loss": 0.9263, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 6.82, | |
| "learning_rate": 3.074188894812617e-05, | |
| "loss": 0.9491, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "learning_rate": 2.7021311283113737e-05, | |
| "loss": 0.9332, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 2.3455651659094773e-05, | |
| "loss": 0.9, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "eval_loss": 0.42447754740715027, | |
| "eval_runtime": 108.0825, | |
| "eval_samples_per_second": 38.979, | |
| "eval_steps_per_second": 0.611, | |
| "eval_wer": 0.4364881192106323, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.18, | |
| "learning_rate": 2.0068949089374516e-05, | |
| "loss": 0.9037, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 7.29, | |
| "learning_rate": 1.688403609211766e-05, | |
| "loss": 0.863, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 7.41, | |
| "learning_rate": 1.3922384757761936e-05, | |
| "loss": 0.8822, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 7.53, | |
| "learning_rate": 1.1203961988182233e-05, | |
| "loss": 0.8838, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "learning_rate": 8.74709488355579e-06, | |
| "loss": 0.8905, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 7.76, | |
| "learning_rate": 6.568347184464973e-06, | |
| "loss": 0.8846, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 7.88, | |
| "learning_rate": 4.682407602242322e-06, | |
| "loss": 0.8865, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 3.1019907904135227e-06, | |
| "loss": 0.8802, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "learning_rate": 1.8377516248706705e-06, | |
| "loss": 0.8641, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 8.24, | |
| "learning_rate": 8.982133706828155e-07, | |
| "loss": 0.8817, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 8.24, | |
| "eval_loss": 0.4170154631137848, | |
| "eval_runtime": 105.9513, | |
| "eval_samples_per_second": 39.764, | |
| "eval_steps_per_second": 0.623, | |
| "eval_wer": 0.4307289569069674, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 8.35, | |
| "learning_rate": 2.8971021982976964e-07, | |
| "loss": 0.8626, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "learning_rate": 1.6344587259148336e-08, | |
| "loss": 0.8632, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 8.59, | |
| "learning_rate": 9.992004054683143e-05, | |
| "loss": 0.8675, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 8.71, | |
| "learning_rate": 9.951987406202364e-05, | |
| "loss": 0.9324, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "learning_rate": 9.878585380625223e-05, | |
| "loss": 0.9278, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 8.94, | |
| "learning_rate": 9.772292840741648e-05, | |
| "loss": 0.9294, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "learning_rate": 9.633826391065235e-05, | |
| "loss": 0.9066, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 9.18, | |
| "learning_rate": 9.46411954661903e-05, | |
| "loss": 0.908, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 9.29, | |
| "learning_rate": 9.264316439351065e-05, | |
| "loss": 0.901, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "learning_rate": 9.035764104609774e-05, | |
| "loss": 0.8884, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "eval_loss": 0.4221749007701874, | |
| "eval_runtime": 105.5743, | |
| "eval_samples_per_second": 39.906, | |
| "eval_steps_per_second": 0.625, | |
| "eval_wer": 0.4426500201369311, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 9.53, | |
| "learning_rate": 8.780003399682484e-05, | |
| "loss": 0.9003, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 9.65, | |
| "learning_rate": 8.498758615622466e-05, | |
| "loss": 0.8945, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 9.76, | |
| "learning_rate": 8.193925852399696e-05, | |
| "loss": 0.9117, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 9.88, | |
| "learning_rate": 7.867560235747892e-05, | |
| "loss": 0.8941, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 7.521862061889445e-05, | |
| "loss": 0.8736, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 10.12, | |
| "learning_rate": 7.159161963547964e-05, | |
| "loss": 0.8359, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 10.24, | |
| "learning_rate": 6.781905197256384e-05, | |
| "loss": 0.8247, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 10.35, | |
| "learning_rate": 6.392635157892686e-05, | |
| "loss": 0.8058, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 10.47, | |
| "learning_rate": 5.993976231585202e-05, | |
| "loss": 0.808, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 10.59, | |
| "learning_rate": 5.588616102589939e-05, | |
| "loss": 0.8199, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 10.59, | |
| "eval_loss": 0.4070005416870117, | |
| "eval_runtime": 104.9334, | |
| "eval_samples_per_second": 40.149, | |
| "eval_steps_per_second": 0.629, | |
| "eval_wer": 0.41913008457511075, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 10.71, | |
| "learning_rate": 5.17928763342375e-05, | |
| "loss": 0.8051, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 10.82, | |
| "learning_rate": 4.768750440414064e-05, | |
| "loss": 0.8283, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 10.94, | |
| "learning_rate": 4.3597722888794156e-05, | |
| "loss": 0.82, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 11.06, | |
| "learning_rate": 3.955110433370994e-05, | |
| "loss": 0.7931, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 11.18, | |
| "learning_rate": 3.5574930287758105e-05, | |
| "loss": 0.7729, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 11.29, | |
| "learning_rate": 3.1696007376044074e-05, | |
| "loss": 0.7711, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 11.41, | |
| "learning_rate": 2.794048657463315e-05, | |
| "loss": 0.7513, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 11.53, | |
| "learning_rate": 2.4333686905538853e-05, | |
| "loss": 0.7546, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 11.65, | |
| "learning_rate": 2.089992474059062e-05, | |
| "loss": 0.7501, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 11.76, | |
| "learning_rate": 1.7662349864982235e-05, | |
| "loss": 0.7504, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 11.76, | |
| "eval_loss": 0.3886597156524658, | |
| "eval_runtime": 105.3872, | |
| "eval_samples_per_second": 39.976, | |
| "eval_steps_per_second": 0.626, | |
| "eval_wer": 0.404430124848973, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 11.88, | |
| "learning_rate": 1.4642789405730323e-05, | |
| "loss": 0.7703, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 1.1861600677247902e-05, | |
| "loss": 0.7348, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 12.12, | |
| "learning_rate": 9.361444478984221e-06, | |
| "loss": 0.7326, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 12.24, | |
| "learning_rate": 7.108696871859649e-06, | |
| "loss": 0.7105, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 12.35, | |
| "learning_rate": 5.1451144446169495e-06, | |
| "loss": 0.7248, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 12.47, | |
| "learning_rate": 3.483935304386882e-06, | |
| "loss": 0.7322, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 12.59, | |
| "learning_rate": 2.1363588116427025e-06, | |
| "loss": 0.7395, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 12.71, | |
| "learning_rate": 1.120097142214388e-06, | |
| "loss": 0.7192, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 12.82, | |
| "learning_rate": 4.214828618490041e-07, | |
| "loss": 0.7141, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 12.94, | |
| "learning_rate": 5.711771314791392e-08, | |
| "loss": 0.7213, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 12.94, | |
| "eval_loss": 0.3896392285823822, | |
| "eval_runtime": 105.3159, | |
| "eval_samples_per_second": 40.003, | |
| "eval_steps_per_second": 0.627, | |
| "eval_wer": 0.40378574305275877, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 13.06, | |
| "learning_rate": 9.997054182189645e-05, | |
| "loss": 0.7318, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 13.18, | |
| "learning_rate": 9.966130926784864e-05, | |
| "loss": 0.7748, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 13.29, | |
| "learning_rate": 9.901726941301295e-05, | |
| "loss": 0.773, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 13.41, | |
| "learning_rate": 9.804276425417875e-05, | |
| "loss": 0.8016, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 13.53, | |
| "learning_rate": 9.67443637236866e-05, | |
| "loss": 0.7831, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 13.65, | |
| "learning_rate": 9.51308213961676e-05, | |
| "loss": 0.8006, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 13.76, | |
| "learning_rate": 9.321301547357316e-05, | |
| "loss": 0.7881, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 13.88, | |
| "learning_rate": 9.100387544636259e-05, | |
| "loss": 0.7888, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 8.851829492528598e-05, | |
| "loss": 0.7791, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 14.12, | |
| "learning_rate": 8.57730312314333e-05, | |
| "loss": 0.7637, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 14.12, | |
| "eval_loss": 0.4040660262107849, | |
| "eval_runtime": 105.4427, | |
| "eval_samples_per_second": 39.955, | |
| "eval_steps_per_second": 0.626, | |
| "eval_wer": 0.408135320177205, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 14.24, | |
| "learning_rate": 8.278659242149581e-05, | |
| "loss": 0.7497, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 14.35, | |
| "learning_rate": 7.957911250989374e-05, | |
| "loss": 0.7491, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 14.47, | |
| "learning_rate": 7.617221572899955e-05, | |
| "loss": 0.7514, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 14.59, | |
| "learning_rate": 7.258887074258988e-05, | |
| "loss": 0.7589, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 14.71, | |
| "learning_rate": 6.88532357953903e-05, | |
| "loss": 0.7158, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 14.82, | |
| "learning_rate": 6.49904958426878e-05, | |
| "loss": 0.7142, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 14.94, | |
| "learning_rate": 6.1026692758051155e-05, | |
| "loss": 0.7198, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 15.06, | |
| "learning_rate": 5.698854976386778e-05, | |
| "loss": 0.7107, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 15.18, | |
| "learning_rate": 5.2903291268354404e-05, | |
| "loss": 0.6709, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 15.29, | |
| "learning_rate": 4.8798459323665145e-05, | |
| "loss": 0.6743, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 15.29, | |
| "eval_loss": 0.4003716707229614, | |
| "eval_runtime": 104.8505, | |
| "eval_samples_per_second": 40.181, | |
| "eval_steps_per_second": 0.629, | |
| "eval_wer": 0.39714055577929924, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 15.41, | |
| "learning_rate": 4.470172794250435e-05, | |
| "loss": 0.7087, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 15.53, | |
| "learning_rate": 4.064071652508525e-05, | |
| "loss": 0.6737, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 15.65, | |
| "learning_rate": 3.6642803654275966e-05, | |
| "loss": 0.6752, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 15.76, | |
| "learning_rate": 3.2734942514291747e-05, | |
| "loss": 0.6622, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 15.88, | |
| "learning_rate": 2.8943479177344868e-05, | |
| "loss": 0.6889, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 2.5293974983331425e-05, | |
| "loss": 0.6654, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 16.12, | |
| "learning_rate": 2.181103421003689e-05, | |
| "loss": 0.6492, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 16.24, | |
| "learning_rate": 1.851813819567597e-05, | |
| "loss": 0.6363, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 16.35, | |
| "learning_rate": 1.543748703208282e-05, | |
| "loss": 0.6505, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 16.47, | |
| "learning_rate": 1.2589849895824684e-05, | |
| "loss": 0.6261, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 16.47, | |
| "eval_loss": 0.38466739654541016, | |
| "eval_runtime": 105.1698, | |
| "eval_samples_per_second": 40.059, | |
| "eval_steps_per_second": 0.628, | |
| "eval_wer": 0.38771647200966575, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 16.59, | |
| "learning_rate": 9.994425026280951e-06, | |
| "loss": 0.6213, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 16.71, | |
| "learning_rate": 7.668710294687454e-06, | |
| "loss": 0.6348, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 16.82, | |
| "learning_rate": 5.6283852367463915e-06, | |
| "loss": 0.6538, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 16.94, | |
| "learning_rate": 3.8872053441162186e-06, | |
| "loss": 0.6341, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 17.06, | |
| "learning_rate": 2.456909327447748e-06, | |
| "loss": 0.6269, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 17.18, | |
| "learning_rate": 1.3471399761826819e-06, | |
| "loss": 0.6414, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 17.29, | |
| "learning_rate": 5.65379148662526e-07, | |
| "loss": 0.6082, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 17.41, | |
| "learning_rate": 1.168973308325827e-07, | |
| "loss": 0.6305, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 17.53, | |
| "learning_rate": 9.999528189639287e-05, | |
| "loss": 0.6196, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 17.65, | |
| "learning_rate": 9.977040224155017e-05, | |
| "loss": 0.6818, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 17.65, | |
| "eval_loss": 0.4097123444080353, | |
| "eval_runtime": 105.0656, | |
| "eval_samples_per_second": 40.099, | |
| "eval_steps_per_second": 0.628, | |
| "eval_wer": 0.3959323399113975, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 17.76, | |
| "learning_rate": 9.920997980140259e-05, | |
| "loss": 0.7001, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 17.88, | |
| "learning_rate": 9.831779283969819e-05, | |
| "loss": 0.7143, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 9.709985631481902e-05, | |
| "loss": 0.6883, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 18.12, | |
| "learning_rate": 9.556438132805179e-05, | |
| "loss": 0.6889, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 18.24, | |
| "learning_rate": 9.372171976587154e-05, | |
| "loss": 0.6823, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 18.35, | |
| "learning_rate": 9.158429450945069e-05, | |
| "loss": 0.6682, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 18.47, | |
| "learning_rate": 8.916651568190705e-05, | |
| "loss": 0.6735, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 18.59, | |
| "learning_rate": 8.648468349793653e-05, | |
| "loss": 0.6835, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 18.71, | |
| "learning_rate": 8.355687837080098e-05, | |
| "loss": 0.684, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 18.82, | |
| "learning_rate": 8.040283901754784e-05, | |
| "loss": 0.7017, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 18.82, | |
| "eval_loss": 0.39238452911376953, | |
| "eval_runtime": 107.5856, | |
| "eval_samples_per_second": 39.16, | |
| "eval_steps_per_second": 0.613, | |
| "eval_wer": 0.3910189287152638, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 18.94, | |
| "learning_rate": 7.704382938425531e-05, | |
| "loss": 0.6519, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 19.06, | |
| "learning_rate": 7.350249528846678e-05, | |
| "loss": 0.6436, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 19.18, | |
| "learning_rate": 6.987810048655369e-05, | |
| "loss": 0.64, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 19.29, | |
| "learning_rate": 6.604723032247485e-05, | |
| "loss": 0.6514, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 19.41, | |
| "learning_rate": 6.210817271934942e-05, | |
| "loss": 0.6311, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 19.53, | |
| "learning_rate": 5.8087484070135967e-05, | |
| "loss": 0.6429, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 19.65, | |
| "learning_rate": 5.4012271109106916e-05, | |
| "loss": 0.6398, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 19.76, | |
| "learning_rate": 4.991000816329353e-05, | |
| "loss": 0.6158, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 19.88, | |
| "learning_rate": 4.580835192568782e-05, | |
| "loss": 0.6142, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 4.173495499896826e-05, | |
| "loss": 0.6221, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.3881571292877197, | |
| "eval_runtime": 104.9541, | |
| "eval_samples_per_second": 40.141, | |
| "eval_steps_per_second": 0.629, | |
| "eval_wer": 0.381876761981474, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 20.12, | |
| "learning_rate": 3.7717279466800834e-05, | |
| "loss": 0.6, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 20.24, | |
| "learning_rate": 3.3782411749583854e-05, | |
| "loss": 0.5918, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 20.35, | |
| "learning_rate": 2.9956879992845265e-05, | |
| "loss": 0.569, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 20.47, | |
| "learning_rate": 2.6266475219424137e-05, | |
| "loss": 0.5599, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 20.59, | |
| "learning_rate": 2.2736077451197112e-05, | |
| "loss": 0.5745, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 20.71, | |
| "learning_rate": 1.9389487972603084e-05, | |
| "loss": 0.5829, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 20.82, | |
| "learning_rate": 1.6249268866814737e-05, | |
| "loss": 0.5823, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 20.94, | |
| "learning_rate": 1.3336590906375096e-05, | |
| "loss": 0.5708, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 21.06, | |
| "learning_rate": 1.067109082379022e-05, | |
| "loss": 0.5556, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 21.18, | |
| "learning_rate": 8.270738924335186e-06, | |
| "loss": 0.5568, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 21.18, | |
| "eval_loss": 0.3854876756668091, | |
| "eval_runtime": 105.4508, | |
| "eval_samples_per_second": 39.952, | |
| "eval_steps_per_second": 0.626, | |
| "eval_wer": 0.3769230769230769, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 21.29, | |
| "learning_rate": 6.151717933601536e-06, | |
| "loss": 0.5469, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 21.41, | |
| "learning_rate": 4.328313896574343e-06, | |
| "loss": 0.5706, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 21.53, | |
| "learning_rate": 2.812819863776944e-06, | |
| "loss": 0.553, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 21.65, | |
| "learning_rate": 1.6154530138129676e-06, | |
| "loss": 0.5454, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 21.76, | |
| "learning_rate": 7.442857710505957e-07, | |
| "loss": 0.5453, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 21.88, | |
| "learning_rate": 2.051913828404417e-07, | |
| "loss": 0.5537, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "learning_rate": 1.8043231753583911e-09, | |
| "loss": 0.554, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 22.12, | |
| "learning_rate": 9.986450421025644e-05, | |
| "loss": 0.5931, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 22.24, | |
| "learning_rate": 9.93946355404309e-05, | |
| "loss": 0.6287, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 22.35, | |
| "learning_rate": 9.859175743443652e-05, | |
| "loss": 0.6087, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 22.35, | |
| "eval_loss": 0.394783079624176, | |
| "eval_runtime": 106.0156, | |
| "eval_samples_per_second": 39.739, | |
| "eval_steps_per_second": 0.623, | |
| "eval_wer": 0.38787756745871926, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 22.47, | |
| "learning_rate": 9.746128274697833e-05, | |
| "loss": 0.6281, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 22.59, | |
| "learning_rate": 9.601083292791942e-05, | |
| "loss": 0.6269, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 22.71, | |
| "learning_rate": 9.425018663988473e-05, | |
| "loss": 0.6171, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 22.82, | |
| "learning_rate": 9.219121383233855e-05, | |
| "loss": 0.6135, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 22.94, | |
| "learning_rate": 8.984779571659518e-05, | |
| "loss": 0.6233, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 23.06, | |
| "learning_rate": 8.723573118127667e-05, | |
| "loss": 0.6153, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 23.18, | |
| "learning_rate": 8.43726302791475e-05, | |
| "loss": 0.6014, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 23.29, | |
| "learning_rate": 8.127779550341711e-05, | |
| "loss": 0.5969, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 23.41, | |
| "learning_rate": 7.797209165392523e-05, | |
| "loss": 0.579, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 23.53, | |
| "learning_rate": 7.451360514395531e-05, | |
| "loss": 0.5808, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 23.53, | |
| "eval_loss": 0.4057287871837616, | |
| "eval_runtime": 104.5213, | |
| "eval_samples_per_second": 40.308, | |
| "eval_steps_per_second": 0.631, | |
| "eval_wer": 0.3873942811115586, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 23.65, | |
| "learning_rate": 7.0855823703674e-05, | |
| "loss": 0.5769, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 23.76, | |
| "learning_rate": 6.70574361830682e-05, | |
| "loss": 0.5719, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 23.88, | |
| "learning_rate": 6.314405060359115e-05, | |
| "loss": 0.5757, | |
| "step": 20300 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 5.91420502821974e-05, | |
| "loss": 0.5741, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 24.12, | |
| "learning_rate": 5.507841595993307e-05, | |
| "loss": 0.5505, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 24.24, | |
| "learning_rate": 5.098054390280248e-05, | |
| "loss": 0.5231, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 24.35, | |
| "learning_rate": 4.691704783544732e-05, | |
| "loss": 0.5522, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 24.47, | |
| "learning_rate": 4.28332788746043e-05, | |
| "loss": 0.5458, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 24.59, | |
| "learning_rate": 3.879782661323741e-05, | |
| "loss": 0.5427, | |
| "step": 20900 | |
| }, | |
| { | |
| "epoch": 24.71, | |
| "learning_rate": 3.483789731914372e-05, | |
| "loss": 0.5413, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 24.71, | |
| "eval_loss": 0.3984732925891876, | |
| "eval_runtime": 107.6668, | |
| "eval_samples_per_second": 39.13, | |
| "eval_steps_per_second": 0.613, | |
| "eval_wer": 0.3810712847362062, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 24.82, | |
| "learning_rate": 3.098018809833621e-05, | |
| "loss": 0.5312, | |
| "step": 21100 | |
| }, | |
| { | |
| "epoch": 24.94, | |
| "learning_rate": 2.7250706908125788e-05, | |
| "loss": 0.5168, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 25.06, | |
| "learning_rate": 2.3674597216315607e-05, | |
| "loss": 0.4865, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 25.18, | |
| "learning_rate": 2.0275968488622116e-05, | |
| "loss": 0.4999, | |
| "step": 21400 | |
| }, | |
| { | |
| "epoch": 25.29, | |
| "learning_rate": 1.707773364714661e-05, | |
| "loss": 0.509, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 25.41, | |
| "learning_rate": 1.4101454595720243e-05, | |
| "loss": 0.4925, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 25.53, | |
| "learning_rate": 1.1367196853562878e-05, | |
| "loss": 0.4885, | |
| "step": 21700 | |
| }, | |
| { | |
| "epoch": 25.65, | |
| "learning_rate": 8.89339427728868e-06, | |
| "loss": 0.4897, | |
| "step": 21800 | |
| }, | |
| { | |
| "epoch": 25.76, | |
| "learning_rate": 6.696724783276548e-06, | |
| "loss": 0.5034, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 25.88, | |
| "learning_rate": 4.791997908263651e-06, | |
| "loss": 0.4731, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 25.88, | |
| "eval_loss": 0.3930116593837738, | |
| "eval_runtime": 106.2837, | |
| "eval_samples_per_second": 39.639, | |
| "eval_steps_per_second": 0.621, | |
| "eval_wer": 0.36939186467982277, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "learning_rate": 3.1920549662057043e-06, | |
| "loss": 0.4943, | |
| "step": 22100 | |
| }, | |
| { | |
| "epoch": 26.12, | |
| "learning_rate": 1.9076824745293208e-06, | |
| "loss": 0.4817, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 26.24, | |
| "learning_rate": 9.475394334397813e-07, | |
| "loss": 0.4809, | |
| "step": 22300 | |
| }, | |
| { | |
| "epoch": 26.35, | |
| "learning_rate": 3.180989485537966e-07, | |
| "loss": 0.478, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 26.47, | |
| "learning_rate": 2.3604590426795636e-08, | |
| "loss": 0.4911, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 26.59, | |
| "learning_rate": 9.993395821480979e-05, | |
| "loss": 0.5161, | |
| "step": 22600 | |
| }, | |
| { | |
| "epoch": 26.71, | |
| "learning_rate": 9.95548755708218e-05, | |
| "loss": 0.5251, | |
| "step": 22700 | |
| }, | |
| { | |
| "epoch": 26.82, | |
| "learning_rate": 9.884170318221471e-05, | |
| "loss": 0.5496, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 26.94, | |
| "learning_rate": 9.779924912444897e-05, | |
| "loss": 0.5606, | |
| "step": 22900 | |
| }, | |
| { | |
| "epoch": 27.06, | |
| "learning_rate": 9.643454142869053e-05, | |
| "loss": 0.5363, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 27.06, | |
| "eval_loss": 0.4152192175388336, | |
| "eval_runtime": 104.8412, | |
| "eval_samples_per_second": 40.185, | |
| "eval_steps_per_second": 0.63, | |
| "eval_wer": 0.3826419653644785, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 27.18, | |
| "learning_rate": 9.47567807001332e-05, | |
| "loss": 0.5215, | |
| "step": 23100 | |
| }, | |
| { | |
| "epoch": 27.29, | |
| "learning_rate": 9.277727808923149e-05, | |
| "loss": 0.5371, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 27.41, | |
| "learning_rate": 9.05093790340303e-05, | |
| "loss": 0.5601, | |
| "step": 23300 | |
| }, | |
| { | |
| "epoch": 27.53, | |
| "learning_rate": 8.796837328770604e-05, | |
| "loss": 0.5388, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 27.65, | |
| "learning_rate": 8.5171391837898e-05, | |
| "loss": 0.5428, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 27.76, | |
| "learning_rate": 8.213729141277887e-05, | |
| "loss": 0.5581, | |
| "step": 23600 | |
| }, | |
| { | |
| "epoch": 27.88, | |
| "learning_rate": 7.888652735250435e-05, | |
| "loss": 0.5678, | |
| "step": 23700 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "learning_rate": 7.544101570311861e-05, | |
| "loss": 0.5225, | |
| "step": 23800 | |
| }, | |
| { | |
| "epoch": 28.12, | |
| "learning_rate": 7.182398546265166e-05, | |
| "loss": 0.5327, | |
| "step": 23900 | |
| }, | |
| { | |
| "epoch": 28.24, | |
| "learning_rate": 6.805982197554043e-05, | |
| "loss": 0.5235, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 28.24, | |
| "eval_loss": 0.40879756212234497, | |
| "eval_runtime": 106.7199, | |
| "eval_samples_per_second": 39.477, | |
| "eval_steps_per_second": 0.618, | |
| "eval_wer": 0.3765606121627064, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 28.35, | |
| "learning_rate": 6.417390253117627e-05, | |
| "loss": 0.5206, | |
| "step": 24100 | |
| }, | |
| { | |
| "epoch": 28.47, | |
| "learning_rate": 6.0192425274944955e-05, | |
| "loss": 0.5239, | |
| "step": 24200 | |
| }, | |
| { | |
| "epoch": 28.59, | |
| "learning_rate": 5.614223258520837e-05, | |
| "loss": 0.4954, | |
| "step": 24300 | |
| }, | |
| { | |
| "epoch": 28.71, | |
| "learning_rate": 5.2050630106986734e-05, | |
| "loss": 0.4954, | |
| "step": 24400 | |
| }, | |
| { | |
| "epoch": 28.82, | |
| "learning_rate": 4.79452026623847e-05, | |
| "loss": 0.4967, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 28.94, | |
| "learning_rate": 4.385362827885607e-05, | |
| "loss": 0.4762, | |
| "step": 24600 | |
| }, | |
| { | |
| "epoch": 29.06, | |
| "learning_rate": 3.980349158909619e-05, | |
| "loss": 0.476, | |
| "step": 24700 | |
| }, | |
| { | |
| "epoch": 29.18, | |
| "learning_rate": 3.582209786058425e-05, | |
| "loss": 0.4622, | |
| "step": 24800 | |
| }, | |
| { | |
| "epoch": 29.29, | |
| "learning_rate": 3.193628890855341e-05, | |
| "loss": 0.4724, | |
| "step": 24900 | |
| }, | |
| { | |
| "epoch": 29.41, | |
| "learning_rate": 2.8172262133470838e-05, | |
| "loss": 0.4951, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 29.41, | |
| "eval_loss": 0.41040393710136414, | |
| "eval_runtime": 105.7062, | |
| "eval_samples_per_second": 39.856, | |
| "eval_steps_per_second": 0.624, | |
| "eval_wer": 0.37140555779299234, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 29.53, | |
| "learning_rate": 2.455539390304083e-05, | |
| "loss": 0.448, | |
| "step": 25100 | |
| }, | |
| { | |
| "epoch": 29.65, | |
| "learning_rate": 2.1110068469458716e-05, | |
| "loss": 0.4546, | |
| "step": 25200 | |
| }, | |
| { | |
| "epoch": 29.76, | |
| "learning_rate": 1.785951357532256e-05, | |
| "loss": 0.4388, | |
| "step": 25300 | |
| }, | |
| { | |
| "epoch": 29.88, | |
| "learning_rate": 1.4825643856517025e-05, | |
| "loss": 0.4521, | |
| "step": 25400 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 1.2028913097818905e-05, | |
| "loss": 0.4474, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 30.12, | |
| "learning_rate": 9.488176337288151e-06, | |
| "loss": 0.431, | |
| "step": 25600 | |
| }, | |
| { | |
| "epoch": 30.24, | |
| "learning_rate": 7.2205627491117756e-06, | |
| "loss": 0.4444, | |
| "step": 25700 | |
| }, | |
| { | |
| "epoch": 30.35, | |
| "learning_rate": 5.2596790812937e-06, | |
| "loss": 0.4279, | |
| "step": 25800 | |
| }, | |
| { | |
| "epoch": 30.47, | |
| "learning_rate": 3.579153928305451e-06, | |
| "loss": 0.4397, | |
| "step": 25900 | |
| }, | |
| { | |
| "epoch": 30.59, | |
| "learning_rate": 2.2115894765690538e-06, | |
| "loss": 0.4422, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 30.59, | |
| "eval_loss": 0.4197729825973511, | |
| "eval_runtime": 105.5173, | |
| "eval_samples_per_second": 39.927, | |
| "eval_steps_per_second": 0.625, | |
| "eval_wer": 0.3667740636327024, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 30.71, | |
| "learning_rate": 1.1662055909953062e-06, | |
| "loss": 0.4333, | |
| "step": 26100 | |
| }, | |
| { | |
| "epoch": 30.82, | |
| "learning_rate": 4.500500551037179e-07, | |
| "loss": 0.4262, | |
| "step": 26200 | |
| }, | |
| { | |
| "epoch": 30.94, | |
| "learning_rate": 6.795105617835584e-08, | |
| "loss": 0.423, | |
| "step": 26300 | |
| }, | |
| { | |
| "epoch": 31.06, | |
| "learning_rate": 9.997751536547195e-05, | |
| "loss": 0.4465, | |
| "step": 26400 | |
| }, | |
| { | |
| "epoch": 31.18, | |
| "learning_rate": 9.96860426836977e-05, | |
| "loss": 0.4738, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 31.29, | |
| "learning_rate": 9.905959595305095e-05, | |
| "loss": 0.5112, | |
| "step": 26600 | |
| }, | |
| { | |
| "epoch": 31.41, | |
| "learning_rate": 9.81023985607534e-05, | |
| "loss": 0.4893, | |
| "step": 26700 | |
| }, | |
| { | |
| "epoch": 31.53, | |
| "learning_rate": 9.682090375340612e-05, | |
| "loss": 0.5042, | |
| "step": 26800 | |
| }, | |
| { | |
| "epoch": 31.65, | |
| "learning_rate": 9.522375113040246e-05, | |
| "loss": 0.5001, | |
| "step": 26900 | |
| }, | |
| { | |
| "epoch": 31.76, | |
| "learning_rate": 9.332170839735737e-05, | |
| "loss": 0.4968, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 31.76, | |
| "eval_loss": 0.4139433801174164, | |
| "eval_runtime": 105.1015, | |
| "eval_samples_per_second": 40.085, | |
| "eval_steps_per_second": 0.628, | |
| "eval_wer": 0.3725734997986307, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 31.88, | |
| "learning_rate": 9.112759877224125e-05, | |
| "loss": 0.5126, | |
| "step": 27100 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "learning_rate": 8.865621453363166e-05, | |
| "loss": 0.527, | |
| "step": 27200 | |
| }, | |
| { | |
| "epoch": 32.12, | |
| "learning_rate": 8.592421729392543e-05, | |
| "loss": 0.4913, | |
| "step": 27300 | |
| }, | |
| { | |
| "epoch": 32.24, | |
| "learning_rate": 8.295002566985007e-05, | |
| "loss": 0.4655, | |
| "step": 27400 | |
| }, | |
| { | |
| "epoch": 32.35, | |
| "learning_rate": 7.975369110757874e-05, | |
| "loss": 0.465, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 32.47, | |
| "learning_rate": 7.635676269961402e-05, | |
| "loss": 0.4849, | |
| "step": 27600 | |
| }, | |
| { | |
| "epoch": 32.59, | |
| "learning_rate": 7.278214190481817e-05, | |
| "loss": 0.4657, | |
| "step": 27700 | |
| }, | |
| { | |
| "epoch": 32.71, | |
| "learning_rate": 6.905392815104131e-05, | |
| "loss": 0.4617, | |
| "step": 27800 | |
| }, | |
| { | |
| "epoch": 32.82, | |
| "learning_rate": 6.51972563612662e-05, | |
| "loss": 0.4816, | |
| "step": 27900 | |
| }, | |
| { | |
| "epoch": 32.94, | |
| "learning_rate": 6.123812749863665e-05, | |
| "loss": 0.4697, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 32.94, | |
| "eval_loss": 0.427847295999527, | |
| "eval_runtime": 105.1717, | |
| "eval_samples_per_second": 40.058, | |
| "eval_steps_per_second": 0.628, | |
| "eval_wer": 0.3773660894079742, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 33.06, | |
| "learning_rate": 5.7203233272805804e-05, | |
| "loss": 0.4574, | |
| "step": 28100 | |
| }, | |
| { | |
| "epoch": 33.18, | |
| "learning_rate": 5.311977618939895e-05, | |
| "loss": 0.4532, | |
| "step": 28200 | |
| }, | |
| { | |
| "epoch": 33.29, | |
| "learning_rate": 4.9015286155785297e-05, | |
| "loss": 0.4403, | |
| "step": 28300 | |
| }, | |
| { | |
| "epoch": 33.41, | |
| "learning_rate": 4.4917434879568456e-05, | |
| "loss": 0.435, | |
| "step": 28400 | |
| }, | |
| { | |
| "epoch": 33.53, | |
| "learning_rate": 4.085384931108528e-05, | |
| "loss": 0.4277, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 33.65, | |
| "learning_rate": 3.6851925387651344e-05, | |
| "loss": 0.4153, | |
| "step": 28600 | |
| }, | |
| { | |
| "epoch": 33.76, | |
| "learning_rate": 3.293864333525176e-05, | |
| "loss": 0.4235, | |
| "step": 28700 | |
| }, | |
| { | |
| "epoch": 33.88, | |
| "learning_rate": 2.914038577288113e-05, | |
| "loss": 0.4215, | |
| "step": 28800 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "learning_rate": 2.5482759845838445e-05, | |
| "loss": 0.4057, | |
| "step": 28900 | |
| }, | |
| { | |
| "epoch": 34.12, | |
| "learning_rate": 2.1990424587120746e-05, | |
| "loss": 0.399, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 34.12, | |
| "eval_loss": 0.4379839301109314, | |
| "eval_runtime": 105.199, | |
| "eval_samples_per_second": 40.048, | |
| "eval_steps_per_second": 0.627, | |
| "eval_wer": 0.3649617398308498, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 34.24, | |
| "learning_rate": 1.868692467081464e-05, | |
| "loss": 0.3893, | |
| "step": 29100 | |
| }, | |
| { | |
| "epoch": 34.35, | |
| "learning_rate": 1.559453167828669e-05, | |
| "loss": 0.404, | |
| "step": 29200 | |
| }, | |
| { | |
| "epoch": 34.47, | |
| "learning_rate": 1.2734093947328829e-05, | |
| "loss": 0.3977, | |
| "step": 29300 | |
| }, | |
| { | |
| "epoch": 34.59, | |
| "learning_rate": 1.0124896016546576e-05, | |
| "loss": 0.4139, | |
| "step": 29400 | |
| }, | |
| { | |
| "epoch": 34.71, | |
| "learning_rate": 7.784528612590087e-06, | |
| "loss": 0.3903, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 34.82, | |
| "learning_rate": 5.728770056750632e-06, | |
| "loss": 0.3923, | |
| "step": 29600 | |
| }, | |
| { | |
| "epoch": 34.94, | |
| "learning_rate": 3.971479890455959e-06, | |
| "loss": 0.3927, | |
| "step": 29700 | |
| }, | |
| { | |
| "epoch": 35.06, | |
| "learning_rate": 2.5245054368224164e-06, | |
| "loss": 0.3755, | |
| "step": 29800 | |
| }, | |
| { | |
| "epoch": 35.18, | |
| "learning_rate": 1.4072598492477495e-06, | |
| "loss": 0.385, | |
| "step": 29900 | |
| }, | |
| { | |
| "epoch": 35.29, | |
| "learning_rate": 6.047175874878519e-07, | |
| "loss": 0.3929, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 35.29, | |
| "eval_loss": 0.441159188747406, | |
| "eval_runtime": 104.6164, | |
| "eval_samples_per_second": 40.271, | |
| "eval_steps_per_second": 0.631, | |
| "eval_wer": 0.3600483286347161, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 35.41, | |
| "learning_rate": 1.3518912298790543e-07, | |
| "loss": 0.3965, | |
| "step": 30100 | |
| }, | |
| { | |
| "epoch": 35.53, | |
| "learning_rate": 9.999816007076456e-05, | |
| "loss": 0.3789, | |
| "step": 30200 | |
| }, | |
| { | |
| "epoch": 35.65, | |
| "learning_rate": 9.979443097833314e-05, | |
| "loss": 0.4195, | |
| "step": 30300 | |
| }, | |
| { | |
| "epoch": 35.76, | |
| "learning_rate": 9.92549971033273e-05, | |
| "loss": 0.4553, | |
| "step": 30400 | |
| }, | |
| { | |
| "epoch": 35.88, | |
| "learning_rate": 9.83834952084967e-05, | |
| "loss": 0.4391, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "learning_rate": 9.718580079735384e-05, | |
| "loss": 0.4444, | |
| "step": 30600 | |
| }, | |
| { | |
| "epoch": 36.12, | |
| "learning_rate": 9.566998850262337e-05, | |
| "loss": 0.4345, | |
| "step": 30700 | |
| }, | |
| { | |
| "epoch": 36.24, | |
| "learning_rate": 9.384627764857165e-05, | |
| "loss": 0.4462, | |
| "step": 30800 | |
| }, | |
| { | |
| "epoch": 36.35, | |
| "learning_rate": 9.172696335422526e-05, | |
| "loss": 0.4393, | |
| "step": 30900 | |
| }, | |
| { | |
| "epoch": 36.47, | |
| "learning_rate": 8.932633364196772e-05, | |
| "loss": 0.4403, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 36.47, | |
| "eval_loss": 0.4456220269203186, | |
| "eval_runtime": 104.0661, | |
| "eval_samples_per_second": 40.484, | |
| "eval_steps_per_second": 0.634, | |
| "eval_wer": 0.3653644784534837, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 36.59, | |
| "learning_rate": 8.66605731103526e-05, | |
| "loss": 0.4542, | |
| "step": 31100 | |
| }, | |
| { | |
| "epoch": 36.71, | |
| "learning_rate": 8.374765382055385e-05, | |
| "loss": 0.4522, | |
| "step": 31200 | |
| }, | |
| { | |
| "epoch": 36.82, | |
| "learning_rate": 8.060721413207527e-05, | |
| "loss": 0.4581, | |
| "step": 31300 | |
| }, | |
| { | |
| "epoch": 36.94, | |
| "learning_rate": 7.726042630458715e-05, | |
| "loss": 0.439, | |
| "step": 31400 | |
| }, | |
| { | |
| "epoch": 37.06, | |
| "learning_rate": 7.372985375849379e-05, | |
| "loss": 0.4288, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 37.18, | |
| "learning_rate": 7.003929895655356e-05, | |
| "loss": 0.4188, | |
| "step": 31600 | |
| }, | |
| { | |
| "epoch": 37.29, | |
| "learning_rate": 6.621364293210756e-05, | |
| "loss": 0.4072, | |
| "step": 31700 | |
| }, | |
| { | |
| "epoch": 37.41, | |
| "learning_rate": 6.227867754578468e-05, | |
| "loss": 0.4275, | |
| "step": 31800 | |
| }, | |
| { | |
| "epoch": 37.53, | |
| "learning_rate": 5.83014302576376e-05, | |
| "loss": 0.4064, | |
| "step": 31900 | |
| }, | |
| { | |
| "epoch": 37.65, | |
| "learning_rate": 5.42284121214101e-05, | |
| "loss": 0.3923, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 37.65, | |
| "eval_loss": 0.4511985182762146, | |
| "eval_runtime": 104.7283, | |
| "eval_samples_per_second": 40.228, | |
| "eval_steps_per_second": 0.63, | |
| "eval_wer": 0.36568666935159083, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 37.76, | |
| "learning_rate": 5.012688681793577e-05, | |
| "loss": 0.3951, | |
| "step": 32100 | |
| }, | |
| { | |
| "epoch": 37.88, | |
| "learning_rate": 4.6024506067159605e-05, | |
| "loss": 0.4101, | |
| "step": 32200 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "learning_rate": 4.19489273562929e-05, | |
| "loss": 0.4005, | |
| "step": 32300 | |
| }, | |
| { | |
| "epoch": 38.12, | |
| "learning_rate": 3.792762747818408e-05, | |
| "loss": 0.3798, | |
| "step": 32400 | |
| }, | |
| { | |
| "epoch": 38.24, | |
| "learning_rate": 3.398771728789724e-05, | |
| "loss": 0.3547, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 38.35, | |
| "learning_rate": 3.0155758926374532e-05, | |
| "loss": 0.3815, | |
| "step": 32600 | |
| }, | |
| { | |
| "epoch": 38.47, | |
| "learning_rate": 2.6457586743428208e-05, | |
| "loss": 0.3657, | |
| "step": 32700 | |
| }, | |
| { | |
| "epoch": 38.59, | |
| "learning_rate": 2.2918133127363645e-05, | |
| "loss": 0.3665, | |
| "step": 32800 | |
| }, | |
| { | |
| "epoch": 38.71, | |
| "learning_rate": 1.9561260415459726e-05, | |
| "loss": 0.3522, | |
| "step": 32900 | |
| }, | |
| { | |
| "epoch": 38.82, | |
| "learning_rate": 1.640960001853372e-05, | |
| "loss": 0.3571, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 38.82, | |
| "eval_loss": 0.45314452052116394, | |
| "eval_runtime": 105.5791, | |
| "eval_samples_per_second": 39.904, | |
| "eval_steps_per_second": 0.625, | |
| "eval_wer": 0.3557792992347966, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 38.94, | |
| "learning_rate": 1.3484399844182983e-05, | |
| "loss": 0.3616, | |
| "step": 33100 | |
| }, | |
| { | |
| "epoch": 39.06, | |
| "learning_rate": 1.0805381047348467e-05, | |
| "loss": 0.3522, | |
| "step": 33200 | |
| }, | |
| { | |
| "epoch": 39.18, | |
| "learning_rate": 8.390605073959684e-06, | |
| "loss": 0.3527, | |
| "step": 33300 | |
| }, | |
| { | |
| "epoch": 39.29, | |
| "learning_rate": 6.256351894029011e-06, | |
| "loss": 0.3405, | |
| "step": 33400 | |
| }, | |
| { | |
| "epoch": 39.41, | |
| "learning_rate": 4.4170102451251805e-06, | |
| "loss": 0.3488, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 39.53, | |
| "learning_rate": 2.884980626183248e-06, | |
| "loss": 0.3519, | |
| "step": 33600 | |
| }, | |
| { | |
| "epoch": 39.65, | |
| "learning_rate": 1.67059169565002e-06, | |
| "loss": 0.3395, | |
| "step": 33700 | |
| }, | |
| { | |
| "epoch": 39.76, | |
| "learning_rate": 7.820306375915687e-07, | |
| "loss": 0.3404, | |
| "step": 33800 | |
| }, | |
| { | |
| "epoch": 39.88, | |
| "learning_rate": 2.2528796522271467e-07, | |
| "loss": 0.3447, | |
| "step": 33900 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "learning_rate": 4.117133982245447e-09, | |
| "loss": 0.3482, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_loss": 0.4571866989135742, | |
| "eval_runtime": 103.9248, | |
| "eval_samples_per_second": 40.539, | |
| "eval_steps_per_second": 0.635, | |
| "eval_wer": 0.3579943616592831, | |
| "step": 34000 | |
| } | |
| ], | |
| "max_steps": 34000, | |
| "num_train_epochs": 40, | |
| "total_flos": 3.007490366928798e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |