| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 30.0, | |
| "global_step": 46650, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 3.6375e-06, | |
| "loss": 9.467, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 7.3875e-06, | |
| "loss": 3.0599, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 1.1137499999999998e-05, | |
| "loss": 2.9476, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 1.48875e-05, | |
| "loss": 2.7387, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 1.86375e-05, | |
| "loss": 2.1172, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_loss": 1.2852152585983276, | |
| "eval_runtime": 410.9823, | |
| "eval_samples_per_second": 20.198, | |
| "eval_steps_per_second": 5.051, | |
| "eval_wer": 0.9782597380064197, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 2.23875e-05, | |
| "loss": 1.7925, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 2.6137499999999995e-05, | |
| "loss": 1.6326, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 2.9887499999999998e-05, | |
| "loss": 1.4856, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 3.36375e-05, | |
| "loss": 1.4719, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 3.7387499999999994e-05, | |
| "loss": 1.4152, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_loss": 0.6433968544006348, | |
| "eval_runtime": 405.9573, | |
| "eval_samples_per_second": 20.448, | |
| "eval_steps_per_second": 5.114, | |
| "eval_wer": 0.6104970937798213, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 4.11375e-05, | |
| "loss": 1.4098, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 4.48875e-05, | |
| "loss": 1.4307, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 4.8637499999999996e-05, | |
| "loss": 1.387, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 5.23875e-05, | |
| "loss": 1.4016, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 5.61375e-05, | |
| "loss": 1.4342, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_loss": 0.4843597710132599, | |
| "eval_runtime": 409.811, | |
| "eval_samples_per_second": 20.256, | |
| "eval_steps_per_second": 5.066, | |
| "eval_wer": 0.3989242647696712, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 5.988749999999999e-05, | |
| "loss": 1.4332, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 6.36375e-05, | |
| "loss": 1.4204, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 6.738749999999999e-05, | |
| "loss": 1.4558, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 7.11375e-05, | |
| "loss": 1.4693, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 7.48875e-05, | |
| "loss": 1.4657, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "eval_loss": 0.5079715251922607, | |
| "eval_runtime": 412.202, | |
| "eval_samples_per_second": 20.138, | |
| "eval_steps_per_second": 5.036, | |
| "eval_wer": 0.44903270582111565, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 7.483706606942889e-05, | |
| "loss": 1.5032, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 7.466909294512878e-05, | |
| "loss": 1.5284, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 7.450111982082866e-05, | |
| "loss": 1.4779, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 7.433314669652855e-05, | |
| "loss": 1.5038, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 7.416517357222843e-05, | |
| "loss": 1.4961, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "eval_loss": 0.4763982892036438, | |
| "eval_runtime": 413.618, | |
| "eval_samples_per_second": 20.069, | |
| "eval_steps_per_second": 5.019, | |
| "eval_wer": 0.4263555131430554, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 7.399720044792832e-05, | |
| "loss": 1.4649, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 7.382922732362821e-05, | |
| "loss": 1.4722, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 7.36612541993281e-05, | |
| "loss": 1.4462, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 7.3493281075028e-05, | |
| "loss": 1.4079, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 7.332530795072789e-05, | |
| "loss": 1.4515, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "eval_loss": 0.4519277513027191, | |
| "eval_runtime": 406.3563, | |
| "eval_samples_per_second": 20.428, | |
| "eval_steps_per_second": 5.109, | |
| "eval_wer": 0.4067840721783638, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 7.315733482642776e-05, | |
| "loss": 1.4643, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 7.298936170212766e-05, | |
| "loss": 1.4414, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 7.282138857782753e-05, | |
| "loss": 1.4453, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 7.265341545352743e-05, | |
| "loss": 1.3962, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 7.248544232922732e-05, | |
| "loss": 1.3924, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "eval_loss": 0.44722867012023926, | |
| "eval_runtime": 405.8333, | |
| "eval_samples_per_second": 20.454, | |
| "eval_steps_per_second": 5.115, | |
| "eval_wer": 0.4132037824238744, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 7.231914893617021e-05, | |
| "loss": 1.455, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 7.215117581187009e-05, | |
| "loss": 1.4377, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 7.198320268756998e-05, | |
| "loss": 1.407, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 7.181522956326987e-05, | |
| "loss": 1.4296, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 7.164725643896976e-05, | |
| "loss": 1.4524, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "eval_loss": 0.4454841911792755, | |
| "eval_runtime": 408.7328, | |
| "eval_samples_per_second": 20.309, | |
| "eval_steps_per_second": 5.079, | |
| "eval_wer": 0.3939273011191117, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 7.147928331466966e-05, | |
| "loss": 1.4244, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 7.131131019036953e-05, | |
| "loss": 1.3799, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 7.114333706606943e-05, | |
| "loss": 1.4029, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 7.09753639417693e-05, | |
| "loss": 1.3899, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 7.08073908174692e-05, | |
| "loss": 1.4328, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "eval_loss": 0.43690094351768494, | |
| "eval_runtime": 407.4846, | |
| "eval_samples_per_second": 20.371, | |
| "eval_steps_per_second": 5.095, | |
| "eval_wer": 0.4068881755877505, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 7.063941769316909e-05, | |
| "loss": 1.4371, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 7.047144456886898e-05, | |
| "loss": 1.3755, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 7.030347144456887e-05, | |
| "loss": 1.384, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 7.013549832026875e-05, | |
| "loss": 1.3929, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 6.996752519596864e-05, | |
| "loss": 1.3456, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "eval_loss": 0.4234367609024048, | |
| "eval_runtime": 406.9215, | |
| "eval_samples_per_second": 20.4, | |
| "eval_steps_per_second": 5.102, | |
| "eval_wer": 0.37740956016309535, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 6.979955207166852e-05, | |
| "loss": 1.363, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 6.963157894736841e-05, | |
| "loss": 1.3519, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 6.94636058230683e-05, | |
| "loss": 1.3668, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 6.92956326987682e-05, | |
| "loss": 1.3869, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 6.912765957446809e-05, | |
| "loss": 1.3725, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "eval_loss": 0.4387490153312683, | |
| "eval_runtime": 404.4269, | |
| "eval_samples_per_second": 20.525, | |
| "eval_steps_per_second": 5.133, | |
| "eval_wer": 0.37891905959920186, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 6.895968645016796e-05, | |
| "loss": 1.3783, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 6.879171332586786e-05, | |
| "loss": 1.3283, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 6.862374020156773e-05, | |
| "loss": 1.3606, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 6.845576707726763e-05, | |
| "loss": 1.3599, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 6.828779395296752e-05, | |
| "loss": 1.3812, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "eval_loss": 0.42978355288505554, | |
| "eval_runtime": 411.3025, | |
| "eval_samples_per_second": 20.182, | |
| "eval_steps_per_second": 5.047, | |
| "eval_wer": 0.3825106272230416, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 6.811982082866741e-05, | |
| "loss": 1.3738, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 6.79518477043673e-05, | |
| "loss": 1.3948, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 6.778387458006718e-05, | |
| "loss": 1.3343, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 6.761590145576707e-05, | |
| "loss": 1.3387, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "learning_rate": 6.744792833146696e-05, | |
| "loss": 1.3282, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "eval_loss": 0.40246763825416565, | |
| "eval_runtime": 404.0105, | |
| "eval_samples_per_second": 20.546, | |
| "eval_steps_per_second": 5.138, | |
| "eval_wer": 0.3702611260518782, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "learning_rate": 6.727995520716684e-05, | |
| "loss": 1.3328, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "learning_rate": 6.711198208286673e-05, | |
| "loss": 1.3629, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 6.694568868980962e-05, | |
| "loss": 1.3236, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 6.677771556550952e-05, | |
| "loss": 1.3341, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 6.66097424412094e-05, | |
| "loss": 1.3326, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "eval_loss": 0.39169707894325256, | |
| "eval_runtime": 403.4284, | |
| "eval_samples_per_second": 20.576, | |
| "eval_steps_per_second": 5.146, | |
| "eval_wer": 0.35023857031317773, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 6.644176931690929e-05, | |
| "loss": 1.3353, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "learning_rate": 6.627379619260918e-05, | |
| "loss": 1.3077, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "learning_rate": 6.610582306830907e-05, | |
| "loss": 1.3383, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 6.593784994400896e-05, | |
| "loss": 1.3256, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "learning_rate": 6.576987681970884e-05, | |
| "loss": 1.3028, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "eval_loss": 0.38892313838005066, | |
| "eval_runtime": 402.0591, | |
| "eval_samples_per_second": 20.646, | |
| "eval_steps_per_second": 5.163, | |
| "eval_wer": 0.3581851305630259, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 6.560190369540873e-05, | |
| "loss": 1.3176, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 6.543393057110861e-05, | |
| "loss": 1.3364, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 6.526763717805152e-05, | |
| "loss": 1.3361, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 6.50996640537514e-05, | |
| "loss": 1.3351, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "learning_rate": 6.493169092945129e-05, | |
| "loss": 1.293, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "eval_loss": 0.38585448265075684, | |
| "eval_runtime": 405.4243, | |
| "eval_samples_per_second": 20.475, | |
| "eval_steps_per_second": 5.121, | |
| "eval_wer": 0.3495965992886267, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 5.21, | |
| "learning_rate": 6.476371780515116e-05, | |
| "loss": 1.3091, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 5.27, | |
| "learning_rate": 6.459742441209406e-05, | |
| "loss": 1.3023, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 5.34, | |
| "learning_rate": 6.442945128779395e-05, | |
| "loss": 1.3101, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 5.4, | |
| "learning_rate": 6.426147816349384e-05, | |
| "loss": 1.3045, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 5.47, | |
| "learning_rate": 6.409350503919372e-05, | |
| "loss": 1.321, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 5.47, | |
| "eval_loss": 0.38752472400665283, | |
| "eval_runtime": 405.1787, | |
| "eval_samples_per_second": 20.487, | |
| "eval_steps_per_second": 5.124, | |
| "eval_wer": 0.35759521124316823, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "learning_rate": 6.392553191489361e-05, | |
| "loss": 1.3002, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "learning_rate": 6.37575587905935e-05, | |
| "loss": 1.3374, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 6.358958566629339e-05, | |
| "loss": 1.3392, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 5.72, | |
| "learning_rate": 6.342161254199327e-05, | |
| "loss": 1.3266, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "learning_rate": 6.325363941769316e-05, | |
| "loss": 1.3165, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "eval_loss": 0.39272764325141907, | |
| "eval_runtime": 404.4461, | |
| "eval_samples_per_second": 20.524, | |
| "eval_steps_per_second": 5.133, | |
| "eval_wer": 0.3589138544287325, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 5.85, | |
| "learning_rate": 6.308566629339305e-05, | |
| "loss": 1.2855, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "learning_rate": 6.291769316909293e-05, | |
| "loss": 1.3208, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 5.98, | |
| "learning_rate": 6.274972004479282e-05, | |
| "loss": 1.2831, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 6.258174692049272e-05, | |
| "loss": 1.2974, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "learning_rate": 6.241377379619261e-05, | |
| "loss": 1.2701, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "eval_loss": 0.4057924151420593, | |
| "eval_runtime": 405.4037, | |
| "eval_samples_per_second": 20.476, | |
| "eval_steps_per_second": 5.121, | |
| "eval_wer": 0.36207165784679446, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "learning_rate": 6.22458006718925e-05, | |
| "loss": 1.305, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "learning_rate": 6.207782754759238e-05, | |
| "loss": 1.2891, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 6.3, | |
| "learning_rate": 6.190985442329227e-05, | |
| "loss": 1.2526, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 6.37, | |
| "learning_rate": 6.174188129899215e-05, | |
| "loss": 1.2761, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "learning_rate": 6.157390817469204e-05, | |
| "loss": 1.2718, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "eval_loss": 0.4211222529411316, | |
| "eval_runtime": 404.2106, | |
| "eval_samples_per_second": 20.536, | |
| "eval_steps_per_second": 5.136, | |
| "eval_wer": 0.39156762383968075, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "learning_rate": 6.140593505039193e-05, | |
| "loss": 1.3058, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 6.123796192609182e-05, | |
| "loss": 1.3029, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 6.106998880179172e-05, | |
| "loss": 1.2677, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 6.69, | |
| "learning_rate": 6.09020156774916e-05, | |
| "loss": 1.2942, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "learning_rate": 6.073404255319148e-05, | |
| "loss": 1.2683, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "eval_loss": 0.3968421220779419, | |
| "eval_runtime": 405.7471, | |
| "eval_samples_per_second": 20.459, | |
| "eval_steps_per_second": 5.116, | |
| "eval_wer": 0.36198490500563896, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 6.82, | |
| "learning_rate": 6.056606942889137e-05, | |
| "loss": 1.2651, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "learning_rate": 6.039809630459126e-05, | |
| "loss": 1.3121, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 6.95, | |
| "learning_rate": 6.0230123180291147e-05, | |
| "loss": 1.2927, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 6.006215005599104e-05, | |
| "loss": 1.3198, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "learning_rate": 5.989417693169092e-05, | |
| "loss": 1.2643, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "eval_loss": 0.4127723276615143, | |
| "eval_runtime": 403.1761, | |
| "eval_samples_per_second": 20.589, | |
| "eval_steps_per_second": 5.149, | |
| "eval_wer": 0.3847662010930858, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "learning_rate": 5.9726203807390815e-05, | |
| "loss": 1.275, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 5.955823068309069e-05, | |
| "loss": 1.263, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "learning_rate": 5.9390257558790585e-05, | |
| "loss": 1.3049, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 7.33, | |
| "learning_rate": 5.922228443449048e-05, | |
| "loss": 1.2993, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 7.4, | |
| "learning_rate": 5.905431131019036e-05, | |
| "loss": 1.2485, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 7.4, | |
| "eval_loss": 0.3849039077758789, | |
| "eval_runtime": 404.3971, | |
| "eval_samples_per_second": 20.527, | |
| "eval_steps_per_second": 5.134, | |
| "eval_wer": 0.37274225730892685, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 7.46, | |
| "learning_rate": 5.888633818589025e-05, | |
| "loss": 1.2486, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 7.52, | |
| "learning_rate": 5.8718365061590145e-05, | |
| "loss": 1.2671, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "learning_rate": 5.855039193729003e-05, | |
| "loss": 1.2393, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "learning_rate": 5.838241881298992e-05, | |
| "loss": 1.2716, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "learning_rate": 5.82144456886898e-05, | |
| "loss": 1.2608, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "eval_loss": 0.3770001530647278, | |
| "eval_runtime": 405.1001, | |
| "eval_samples_per_second": 20.491, | |
| "eval_steps_per_second": 5.125, | |
| "eval_wer": 0.34742777825973803, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "learning_rate": 5.804647256438969e-05, | |
| "loss": 1.2683, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 7.85, | |
| "learning_rate": 5.787849944008958e-05, | |
| "loss": 1.2275, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "learning_rate": 5.771052631578947e-05, | |
| "loss": 1.2302, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 7.97, | |
| "learning_rate": 5.754255319148936e-05, | |
| "loss": 1.2587, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "learning_rate": 5.7376259798432246e-05, | |
| "loss": 1.2388, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "eval_loss": 0.3774464726448059, | |
| "eval_runtime": 408.3719, | |
| "eval_samples_per_second": 20.327, | |
| "eval_steps_per_second": 5.084, | |
| "eval_wer": 0.3574390561290882, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "learning_rate": 5.720828667413213e-05, | |
| "loss": 1.2268, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 8.17, | |
| "learning_rate": 5.704031354983202e-05, | |
| "loss": 1.2256, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 8.23, | |
| "learning_rate": 5.6872340425531914e-05, | |
| "loss": 1.2279, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 8.3, | |
| "learning_rate": 5.67043673012318e-05, | |
| "loss": 1.2701, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 8.36, | |
| "learning_rate": 5.6536394176931684e-05, | |
| "loss": 1.2524, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 8.36, | |
| "eval_loss": 0.3789218068122864, | |
| "eval_runtime": 410.2058, | |
| "eval_samples_per_second": 20.236, | |
| "eval_steps_per_second": 5.061, | |
| "eval_wer": 0.3549926260085018, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 8.42, | |
| "learning_rate": 5.636842105263157e-05, | |
| "loss": 1.2633, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "learning_rate": 5.620044792833146e-05, | |
| "loss": 1.2396, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "learning_rate": 5.6032474804031346e-05, | |
| "loss": 1.2327, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 8.62, | |
| "learning_rate": 5.586450167973124e-05, | |
| "loss": 1.2186, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 8.68, | |
| "learning_rate": 5.569652855543113e-05, | |
| "loss": 1.2458, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 8.68, | |
| "eval_loss": 0.3770112097263336, | |
| "eval_runtime": 406.7819, | |
| "eval_samples_per_second": 20.407, | |
| "eval_steps_per_second": 5.103, | |
| "eval_wer": 0.34097336687776525, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "learning_rate": 5.5528555431131014e-05, | |
| "loss": 1.2264, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 8.81, | |
| "learning_rate": 5.5360582306830906e-05, | |
| "loss": 1.2368, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "learning_rate": 5.5192609182530784e-05, | |
| "loss": 1.2081, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 8.94, | |
| "learning_rate": 5.5024636058230676e-05, | |
| "loss": 1.1874, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 5.485666293393057e-05, | |
| "loss": 1.2505, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.36382830142974854, | |
| "eval_runtime": 407.7152, | |
| "eval_samples_per_second": 20.36, | |
| "eval_steps_per_second": 5.092, | |
| "eval_wer": 0.3403140452849831, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "learning_rate": 5.468868980963045e-05, | |
| "loss": 1.2039, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 9.13, | |
| "learning_rate": 5.4520716685330344e-05, | |
| "loss": 1.2164, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 9.2, | |
| "learning_rate": 5.435274356103023e-05, | |
| "loss": 1.2007, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 9.26, | |
| "learning_rate": 5.418477043673012e-05, | |
| "loss": 1.2053, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 9.32, | |
| "learning_rate": 5.401679731243001e-05, | |
| "loss": 1.2254, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 9.32, | |
| "eval_loss": 0.3770280182361603, | |
| "eval_runtime": 407.5335, | |
| "eval_samples_per_second": 20.369, | |
| "eval_steps_per_second": 5.094, | |
| "eval_wer": 0.35091524247419104, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 9.39, | |
| "learning_rate": 5.384882418812989e-05, | |
| "loss": 1.2297, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 9.45, | |
| "learning_rate": 5.368085106382978e-05, | |
| "loss": 1.2177, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 5.351287793952967e-05, | |
| "loss": 1.2229, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 9.58, | |
| "learning_rate": 5.334490481522956e-05, | |
| "loss": 1.214, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 9.65, | |
| "learning_rate": 5.3176931690929444e-05, | |
| "loss": 1.2459, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 9.65, | |
| "eval_loss": 0.3592114746570587, | |
| "eval_runtime": 405.3152, | |
| "eval_samples_per_second": 20.48, | |
| "eval_steps_per_second": 5.122, | |
| "eval_wer": 0.3349006679968769, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 9.71, | |
| "learning_rate": 5.3008958566629336e-05, | |
| "loss": 1.1947, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 9.77, | |
| "learning_rate": 5.284098544232923e-05, | |
| "loss": 1.2029, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 9.84, | |
| "learning_rate": 5.267301231802911e-05, | |
| "loss": 1.2047, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 9.9, | |
| "learning_rate": 5.2505039193729e-05, | |
| "loss": 1.2219, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 9.97, | |
| "learning_rate": 5.233706606942888e-05, | |
| "loss": 1.2049, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 9.97, | |
| "eval_loss": 0.3599851429462433, | |
| "eval_runtime": 407.0649, | |
| "eval_samples_per_second": 20.392, | |
| "eval_steps_per_second": 5.1, | |
| "eval_wer": 0.3428472282467251, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 10.03, | |
| "learning_rate": 5.2169092945128775e-05, | |
| "loss": 1.1814, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 10.1, | |
| "learning_rate": 5.200111982082866e-05, | |
| "loss": 1.2202, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 10.16, | |
| "learning_rate": 5.183314669652855e-05, | |
| "loss": 1.198, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 10.23, | |
| "learning_rate": 5.166517357222844e-05, | |
| "loss": 1.1967, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 10.29, | |
| "learning_rate": 5.149720044792833e-05, | |
| "loss": 1.2097, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 10.29, | |
| "eval_loss": 0.3626026511192322, | |
| "eval_runtime": 404.0166, | |
| "eval_samples_per_second": 20.546, | |
| "eval_steps_per_second": 5.138, | |
| "eval_wer": 0.3346577600416414, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 10.35, | |
| "learning_rate": 5.132922732362821e-05, | |
| "loss": 1.1901, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 10.42, | |
| "learning_rate": 5.11612541993281e-05, | |
| "loss": 1.2318, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 10.48, | |
| "learning_rate": 5.099328107502799e-05, | |
| "loss": 1.2002, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 10.55, | |
| "learning_rate": 5.082530795072788e-05, | |
| "loss": 1.1804, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 10.61, | |
| "learning_rate": 5.0657334826427766e-05, | |
| "loss": 1.1988, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 10.61, | |
| "eval_loss": 0.37404394149780273, | |
| "eval_runtime": 411.2712, | |
| "eval_samples_per_second": 20.184, | |
| "eval_steps_per_second": 5.048, | |
| "eval_wer": 0.32686735490587315, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 10.68, | |
| "learning_rate": 5.048936170212766e-05, | |
| "loss": 1.1649, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 10.74, | |
| "learning_rate": 5.032138857782754e-05, | |
| "loss": 1.1537, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 10.8, | |
| "learning_rate": 5.0153415453527435e-05, | |
| "loss": 1.187, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 10.87, | |
| "learning_rate": 4.998544232922731e-05, | |
| "loss": 1.1899, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 10.93, | |
| "learning_rate": 4.9817469204927205e-05, | |
| "loss": 1.1671, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 10.93, | |
| "eval_loss": 0.354811429977417, | |
| "eval_runtime": 406.8841, | |
| "eval_samples_per_second": 20.401, | |
| "eval_steps_per_second": 5.102, | |
| "eval_wer": 0.3244903270582112, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 4.9649496080627096e-05, | |
| "loss": 1.1581, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 11.06, | |
| "learning_rate": 4.948152295632698e-05, | |
| "loss": 1.1407, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 11.13, | |
| "learning_rate": 4.931354983202687e-05, | |
| "loss": 1.1716, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 11.19, | |
| "learning_rate": 4.914725643896976e-05, | |
| "loss": 1.1498, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 11.25, | |
| "learning_rate": 4.897928331466965e-05, | |
| "loss": 1.1532, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 11.25, | |
| "eval_loss": 0.33941924571990967, | |
| "eval_runtime": 408.3389, | |
| "eval_samples_per_second": 20.329, | |
| "eval_steps_per_second": 5.084, | |
| "eval_wer": 0.3140279344148521, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 11.32, | |
| "learning_rate": 4.8811310190369536e-05, | |
| "loss": 1.1753, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 11.38, | |
| "learning_rate": 4.864333706606943e-05, | |
| "loss": 1.1516, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 11.45, | |
| "learning_rate": 4.847536394176932e-05, | |
| "loss": 1.1552, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 11.51, | |
| "learning_rate": 4.8307390817469204e-05, | |
| "loss": 1.1669, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 11.58, | |
| "learning_rate": 4.814109742441209e-05, | |
| "loss": 1.1459, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 11.58, | |
| "eval_loss": 0.334937185049057, | |
| "eval_runtime": 408.3377, | |
| "eval_samples_per_second": 20.329, | |
| "eval_steps_per_second": 5.084, | |
| "eval_wer": 0.3156241866921142, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 11.64, | |
| "learning_rate": 4.797312430011198e-05, | |
| "loss": 1.1354, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 11.7, | |
| "learning_rate": 4.780515117581186e-05, | |
| "loss": 1.1545, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 11.77, | |
| "learning_rate": 4.763717805151175e-05, | |
| "loss": 1.1617, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 11.83, | |
| "learning_rate": 4.746920492721164e-05, | |
| "loss": 1.1684, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 11.9, | |
| "learning_rate": 4.730123180291153e-05, | |
| "loss": 1.1511, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 11.9, | |
| "eval_loss": 0.3271713852882385, | |
| "eval_runtime": 407.7735, | |
| "eval_samples_per_second": 20.357, | |
| "eval_steps_per_second": 5.091, | |
| "eval_wer": 0.31099158497440793, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 11.96, | |
| "learning_rate": 4.713325867861142e-05, | |
| "loss": 1.1699, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 12.03, | |
| "learning_rate": 4.6965285554311305e-05, | |
| "loss": 1.149, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 12.09, | |
| "learning_rate": 4.6797312430011196e-05, | |
| "loss": 1.1371, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 12.15, | |
| "learning_rate": 4.662933930571109e-05, | |
| "loss": 1.1382, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 12.22, | |
| "learning_rate": 4.6461366181410966e-05, | |
| "loss": 1.1465, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 12.22, | |
| "eval_loss": 0.33484965562820435, | |
| "eval_runtime": 405.6959, | |
| "eval_samples_per_second": 20.461, | |
| "eval_steps_per_second": 5.117, | |
| "eval_wer": 0.3083716491715104, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 12.28, | |
| "learning_rate": 4.629339305711086e-05, | |
| "loss": 1.1579, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 12.35, | |
| "learning_rate": 4.612541993281074e-05, | |
| "loss": 1.1425, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 12.41, | |
| "learning_rate": 4.5957446808510635e-05, | |
| "loss": 1.1202, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 12.48, | |
| "learning_rate": 4.578947368421052e-05, | |
| "loss": 1.145, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 12.54, | |
| "learning_rate": 4.562150055991041e-05, | |
| "loss": 1.1426, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 12.54, | |
| "eval_loss": 0.3193301260471344, | |
| "eval_runtime": 408.2738, | |
| "eval_samples_per_second": 20.332, | |
| "eval_steps_per_second": 5.085, | |
| "eval_wer": 0.30273271449639977, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 12.6, | |
| "learning_rate": 4.54535274356103e-05, | |
| "loss": 1.1236, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 12.67, | |
| "learning_rate": 4.528555431131019e-05, | |
| "loss": 1.152, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 12.73, | |
| "learning_rate": 4.511758118701007e-05, | |
| "loss": 1.1421, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "learning_rate": 4.494960806270996e-05, | |
| "loss": 1.1262, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 12.86, | |
| "learning_rate": 4.478163493840985e-05, | |
| "loss": 1.1278, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 12.86, | |
| "eval_loss": 0.3318159580230713, | |
| "eval_runtime": 408.3452, | |
| "eval_samples_per_second": 20.328, | |
| "eval_steps_per_second": 5.084, | |
| "eval_wer": 0.3021254446083109, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 12.93, | |
| "learning_rate": 4.461366181410974e-05, | |
| "loss": 1.1444, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "learning_rate": 4.4445688689809627e-05, | |
| "loss": 1.136, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 13.05, | |
| "learning_rate": 4.427771556550952e-05, | |
| "loss": 1.1128, | |
| "step": 20300 | |
| }, | |
| { | |
| "epoch": 13.12, | |
| "learning_rate": 4.41097424412094e-05, | |
| "loss": 1.143, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 13.18, | |
| "learning_rate": 4.394176931690929e-05, | |
| "loss": 1.149, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 13.18, | |
| "eval_loss": 0.31692153215408325, | |
| "eval_runtime": 408.6603, | |
| "eval_samples_per_second": 20.313, | |
| "eval_steps_per_second": 5.08, | |
| "eval_wer": 0.294699401405396, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 13.25, | |
| "learning_rate": 4.377379619260917e-05, | |
| "loss": 1.1399, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 13.31, | |
| "learning_rate": 4.3605823068309065e-05, | |
| "loss": 1.1137, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 13.38, | |
| "learning_rate": 4.343784994400896e-05, | |
| "loss": 1.1364, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 13.44, | |
| "learning_rate": 4.326987681970884e-05, | |
| "loss": 1.118, | |
| "step": 20900 | |
| }, | |
| { | |
| "epoch": 13.5, | |
| "learning_rate": 4.310190369540873e-05, | |
| "loss": 1.114, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 13.5, | |
| "eval_loss": 0.3224009573459625, | |
| "eval_runtime": 407.3904, | |
| "eval_samples_per_second": 20.376, | |
| "eval_steps_per_second": 5.096, | |
| "eval_wer": 0.2986206298256268, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 13.57, | |
| "learning_rate": 4.293393057110862e-05, | |
| "loss": 1.1474, | |
| "step": 21100 | |
| }, | |
| { | |
| "epoch": 13.63, | |
| "learning_rate": 4.276595744680851e-05, | |
| "loss": 1.1221, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 13.7, | |
| "learning_rate": 4.259798432250839e-05, | |
| "loss": 1.1045, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 13.76, | |
| "learning_rate": 4.243001119820828e-05, | |
| "loss": 1.1222, | |
| "step": 21400 | |
| }, | |
| { | |
| "epoch": 13.83, | |
| "learning_rate": 4.226203807390817e-05, | |
| "loss": 1.1249, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 13.83, | |
| "eval_loss": 0.3227294385433197, | |
| "eval_runtime": 406.589, | |
| "eval_samples_per_second": 20.416, | |
| "eval_steps_per_second": 5.106, | |
| "eval_wer": 0.2920621150342674, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 13.89, | |
| "learning_rate": 4.209406494960806e-05, | |
| "loss": 1.0967, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 13.95, | |
| "learning_rate": 4.192609182530795e-05, | |
| "loss": 1.1091, | |
| "step": 21700 | |
| }, | |
| { | |
| "epoch": 14.02, | |
| "learning_rate": 4.175811870100783e-05, | |
| "loss": 1.1065, | |
| "step": 21800 | |
| }, | |
| { | |
| "epoch": 14.08, | |
| "learning_rate": 4.1590145576707725e-05, | |
| "loss": 1.0715, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 14.15, | |
| "learning_rate": 4.142217245240762e-05, | |
| "loss": 1.0968, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 14.15, | |
| "eval_loss": 0.3032504618167877, | |
| "eval_runtime": 408.3356, | |
| "eval_samples_per_second": 20.329, | |
| "eval_steps_per_second": 5.084, | |
| "eval_wer": 0.2877591741129522, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 14.21, | |
| "learning_rate": 4.1254199328107495e-05, | |
| "loss": 1.0806, | |
| "step": 22100 | |
| }, | |
| { | |
| "epoch": 14.28, | |
| "learning_rate": 4.108622620380739e-05, | |
| "loss": 1.0957, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 14.34, | |
| "learning_rate": 4.091825307950727e-05, | |
| "loss": 1.0859, | |
| "step": 22300 | |
| }, | |
| { | |
| "epoch": 14.41, | |
| "learning_rate": 4.0750279955207163e-05, | |
| "loss": 1.1004, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 14.47, | |
| "learning_rate": 4.0582306830907055e-05, | |
| "loss": 1.0851, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 14.47, | |
| "eval_loss": 0.2996005415916443, | |
| "eval_runtime": 404.5618, | |
| "eval_samples_per_second": 20.518, | |
| "eval_steps_per_second": 5.131, | |
| "eval_wer": 0.286301726381539, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 14.53, | |
| "learning_rate": 4.041433370660694e-05, | |
| "loss": 1.1102, | |
| "step": 22600 | |
| }, | |
| { | |
| "epoch": 14.6, | |
| "learning_rate": 4.024636058230683e-05, | |
| "loss": 1.0519, | |
| "step": 22700 | |
| }, | |
| { | |
| "epoch": 14.66, | |
| "learning_rate": 4.007838745800672e-05, | |
| "loss": 1.0832, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 14.73, | |
| "learning_rate": 3.99104143337066e-05, | |
| "loss": 1.0988, | |
| "step": 22900 | |
| }, | |
| { | |
| "epoch": 14.79, | |
| "learning_rate": 3.9744120940649494e-05, | |
| "loss": 1.0985, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 14.79, | |
| "eval_loss": 0.3011046350002289, | |
| "eval_runtime": 408.8862, | |
| "eval_samples_per_second": 20.301, | |
| "eval_steps_per_second": 5.077, | |
| "eval_wer": 0.2843064110349614, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 14.86, | |
| "learning_rate": 3.957614781634938e-05, | |
| "loss": 1.125, | |
| "step": 23100 | |
| }, | |
| { | |
| "epoch": 14.92, | |
| "learning_rate": 3.9408174692049264e-05, | |
| "loss": 1.0515, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 14.98, | |
| "learning_rate": 3.9240201567749156e-05, | |
| "loss": 1.0937, | |
| "step": 23300 | |
| }, | |
| { | |
| "epoch": 15.05, | |
| "learning_rate": 3.907222844344904e-05, | |
| "loss": 1.0683, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 15.11, | |
| "learning_rate": 3.890425531914893e-05, | |
| "loss": 1.0808, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 15.11, | |
| "eval_loss": 0.29318636655807495, | |
| "eval_runtime": 407.4603, | |
| "eval_samples_per_second": 20.373, | |
| "eval_steps_per_second": 5.095, | |
| "eval_wer": 0.2759434371475666, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 15.18, | |
| "learning_rate": 3.8736282194848824e-05, | |
| "loss": 1.0844, | |
| "step": 23600 | |
| }, | |
| { | |
| "epoch": 15.24, | |
| "learning_rate": 3.856830907054871e-05, | |
| "loss": 1.0802, | |
| "step": 23700 | |
| }, | |
| { | |
| "epoch": 15.31, | |
| "learning_rate": 3.84003359462486e-05, | |
| "loss": 1.0849, | |
| "step": 23800 | |
| }, | |
| { | |
| "epoch": 15.37, | |
| "learning_rate": 3.823236282194848e-05, | |
| "loss": 1.0723, | |
| "step": 23900 | |
| }, | |
| { | |
| "epoch": 15.43, | |
| "learning_rate": 3.806438969764837e-05, | |
| "loss": 1.069, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 15.43, | |
| "eval_loss": 0.291872501373291, | |
| "eval_runtime": 411.7858, | |
| "eval_samples_per_second": 20.159, | |
| "eval_steps_per_second": 5.041, | |
| "eval_wer": 0.27495445475839336, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 15.5, | |
| "learning_rate": 3.789641657334826e-05, | |
| "loss": 1.0487, | |
| "step": 24100 | |
| }, | |
| { | |
| "epoch": 15.56, | |
| "learning_rate": 3.772844344904815e-05, | |
| "loss": 1.0635, | |
| "step": 24200 | |
| }, | |
| { | |
| "epoch": 15.63, | |
| "learning_rate": 3.756047032474804e-05, | |
| "loss": 1.0716, | |
| "step": 24300 | |
| }, | |
| { | |
| "epoch": 15.69, | |
| "learning_rate": 3.7392497200447924e-05, | |
| "loss": 1.061, | |
| "step": 24400 | |
| }, | |
| { | |
| "epoch": 15.76, | |
| "learning_rate": 3.722452407614781e-05, | |
| "loss": 1.0602, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 15.76, | |
| "eval_loss": 0.29594287276268005, | |
| "eval_runtime": 406.2825, | |
| "eval_samples_per_second": 20.432, | |
| "eval_steps_per_second": 5.11, | |
| "eval_wer": 0.2712761342933981, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 15.82, | |
| "learning_rate": 3.70565509518477e-05, | |
| "loss": 1.0652, | |
| "step": 24600 | |
| }, | |
| { | |
| "epoch": 15.88, | |
| "learning_rate": 3.688857782754759e-05, | |
| "loss": 1.0604, | |
| "step": 24700 | |
| }, | |
| { | |
| "epoch": 15.95, | |
| "learning_rate": 3.672060470324748e-05, | |
| "loss": 1.0413, | |
| "step": 24800 | |
| }, | |
| { | |
| "epoch": 16.01, | |
| "learning_rate": 3.655263157894736e-05, | |
| "loss": 1.0698, | |
| "step": 24900 | |
| }, | |
| { | |
| "epoch": 16.08, | |
| "learning_rate": 3.6386338185890255e-05, | |
| "loss": 1.0369, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 16.08, | |
| "eval_loss": 0.29310595989227295, | |
| "eval_runtime": 407.0834, | |
| "eval_samples_per_second": 20.391, | |
| "eval_steps_per_second": 5.1, | |
| "eval_wer": 0.2753882189641711, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 16.14, | |
| "learning_rate": 3.621836506159014e-05, | |
| "loss": 1.068, | |
| "step": 25100 | |
| }, | |
| { | |
| "epoch": 16.21, | |
| "learning_rate": 3.605039193729003e-05, | |
| "loss": 1.0281, | |
| "step": 25200 | |
| }, | |
| { | |
| "epoch": 16.27, | |
| "learning_rate": 3.588241881298992e-05, | |
| "loss": 1.0502, | |
| "step": 25300 | |
| }, | |
| { | |
| "epoch": 16.33, | |
| "learning_rate": 3.571444568868981e-05, | |
| "loss": 1.0277, | |
| "step": 25400 | |
| }, | |
| { | |
| "epoch": 16.4, | |
| "learning_rate": 3.5546472564389694e-05, | |
| "loss": 1.0573, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 16.4, | |
| "eval_loss": 0.29200342297554016, | |
| "eval_runtime": 407.2763, | |
| "eval_samples_per_second": 20.382, | |
| "eval_steps_per_second": 5.097, | |
| "eval_wer": 0.27223041554610916, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 16.46, | |
| "learning_rate": 3.537849944008958e-05, | |
| "loss": 1.0411, | |
| "step": 25600 | |
| }, | |
| { | |
| "epoch": 16.53, | |
| "learning_rate": 3.521052631578947e-05, | |
| "loss": 1.0716, | |
| "step": 25700 | |
| }, | |
| { | |
| "epoch": 16.59, | |
| "learning_rate": 3.504255319148936e-05, | |
| "loss": 1.0336, | |
| "step": 25800 | |
| }, | |
| { | |
| "epoch": 16.66, | |
| "learning_rate": 3.487458006718925e-05, | |
| "loss": 1.0523, | |
| "step": 25900 | |
| }, | |
| { | |
| "epoch": 16.72, | |
| "learning_rate": 3.470660694288913e-05, | |
| "loss": 1.051, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 16.72, | |
| "eval_loss": 0.2854558825492859, | |
| "eval_runtime": 411.0951, | |
| "eval_samples_per_second": 20.192, | |
| "eval_steps_per_second": 5.05, | |
| "eval_wer": 0.26317341892946994, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 16.78, | |
| "learning_rate": 3.4538633818589024e-05, | |
| "loss": 1.031, | |
| "step": 26100 | |
| }, | |
| { | |
| "epoch": 16.85, | |
| "learning_rate": 3.4370660694288916e-05, | |
| "loss": 1.0392, | |
| "step": 26200 | |
| }, | |
| { | |
| "epoch": 16.91, | |
| "learning_rate": 3.42026875699888e-05, | |
| "loss": 1.0446, | |
| "step": 26300 | |
| }, | |
| { | |
| "epoch": 16.98, | |
| "learning_rate": 3.4036394176931686e-05, | |
| "loss": 1.0524, | |
| "step": 26400 | |
| }, | |
| { | |
| "epoch": 17.04, | |
| "learning_rate": 3.386842105263158e-05, | |
| "loss": 1.0279, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 17.04, | |
| "eval_loss": 0.2849719524383545, | |
| "eval_runtime": 410.534, | |
| "eval_samples_per_second": 20.22, | |
| "eval_steps_per_second": 5.057, | |
| "eval_wer": 0.2649084757525809, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 17.11, | |
| "learning_rate": 3.370044792833146e-05, | |
| "loss": 1.0467, | |
| "step": 26600 | |
| }, | |
| { | |
| "epoch": 17.17, | |
| "learning_rate": 3.3532474804031355e-05, | |
| "loss": 1.0317, | |
| "step": 26700 | |
| }, | |
| { | |
| "epoch": 17.23, | |
| "learning_rate": 3.336450167973124e-05, | |
| "loss": 1.0061, | |
| "step": 26800 | |
| }, | |
| { | |
| "epoch": 17.3, | |
| "learning_rate": 3.3196528555431125e-05, | |
| "loss": 0.9985, | |
| "step": 26900 | |
| }, | |
| { | |
| "epoch": 17.36, | |
| "learning_rate": 3.3028555431131016e-05, | |
| "loss": 1.0496, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 17.36, | |
| "eval_loss": 0.2816503047943115, | |
| "eval_runtime": 410.4293, | |
| "eval_samples_per_second": 20.225, | |
| "eval_steps_per_second": 5.058, | |
| "eval_wer": 0.25845406437060814, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 17.43, | |
| "learning_rate": 3.28605823068309e-05, | |
| "loss": 1.0298, | |
| "step": 27100 | |
| }, | |
| { | |
| "epoch": 17.49, | |
| "learning_rate": 3.269260918253079e-05, | |
| "loss": 1.024, | |
| "step": 27200 | |
| }, | |
| { | |
| "epoch": 17.56, | |
| "learning_rate": 3.252463605823068e-05, | |
| "loss": 1.0382, | |
| "step": 27300 | |
| }, | |
| { | |
| "epoch": 17.62, | |
| "learning_rate": 3.235666293393057e-05, | |
| "loss": 1.0334, | |
| "step": 27400 | |
| }, | |
| { | |
| "epoch": 17.68, | |
| "learning_rate": 3.2188689809630455e-05, | |
| "loss": 1.0516, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 17.68, | |
| "eval_loss": 0.2960517704486847, | |
| "eval_runtime": 411.8118, | |
| "eval_samples_per_second": 20.157, | |
| "eval_steps_per_second": 5.041, | |
| "eval_wer": 0.263503079725861, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 17.75, | |
| "learning_rate": 3.2020716685330346e-05, | |
| "loss": 1.0131, | |
| "step": 27600 | |
| }, | |
| { | |
| "epoch": 17.81, | |
| "learning_rate": 3.185274356103023e-05, | |
| "loss": 1.0252, | |
| "step": 27700 | |
| }, | |
| { | |
| "epoch": 17.88, | |
| "learning_rate": 3.168477043673012e-05, | |
| "loss": 1.0388, | |
| "step": 27800 | |
| }, | |
| { | |
| "epoch": 17.94, | |
| "learning_rate": 3.151679731243001e-05, | |
| "loss": 1.0139, | |
| "step": 27900 | |
| }, | |
| { | |
| "epoch": 18.01, | |
| "learning_rate": 3.13488241881299e-05, | |
| "loss": 1.0244, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 18.01, | |
| "eval_loss": 0.2780766785144806, | |
| "eval_runtime": 409.1826, | |
| "eval_samples_per_second": 20.287, | |
| "eval_steps_per_second": 5.074, | |
| "eval_wer": 0.2589398802810792, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 18.07, | |
| "learning_rate": 3.1180851063829785e-05, | |
| "loss": 1.0062, | |
| "step": 28100 | |
| }, | |
| { | |
| "epoch": 18.14, | |
| "learning_rate": 3.101287793952967e-05, | |
| "loss": 0.9811, | |
| "step": 28200 | |
| }, | |
| { | |
| "epoch": 18.2, | |
| "learning_rate": 3.084490481522956e-05, | |
| "loss": 0.9957, | |
| "step": 28300 | |
| }, | |
| { | |
| "epoch": 18.26, | |
| "learning_rate": 3.067693169092945e-05, | |
| "loss": 0.9841, | |
| "step": 28400 | |
| }, | |
| { | |
| "epoch": 18.33, | |
| "learning_rate": 3.0508958566629335e-05, | |
| "loss": 1.0099, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 18.33, | |
| "eval_loss": 0.2782675623893738, | |
| "eval_runtime": 409.0441, | |
| "eval_samples_per_second": 20.294, | |
| "eval_steps_per_second": 5.075, | |
| "eval_wer": 0.25651080072872384, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 18.39, | |
| "learning_rate": 3.0342665173572227e-05, | |
| "loss": 0.9978, | |
| "step": 28600 | |
| }, | |
| { | |
| "epoch": 18.46, | |
| "learning_rate": 3.0174692049272116e-05, | |
| "loss": 1.0051, | |
| "step": 28700 | |
| }, | |
| { | |
| "epoch": 18.52, | |
| "learning_rate": 3.0006718924972e-05, | |
| "loss": 1.0229, | |
| "step": 28800 | |
| }, | |
| { | |
| "epoch": 18.59, | |
| "learning_rate": 2.983874580067189e-05, | |
| "loss": 0.9918, | |
| "step": 28900 | |
| }, | |
| { | |
| "epoch": 18.65, | |
| "learning_rate": 2.967077267637178e-05, | |
| "loss": 1.0016, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 18.65, | |
| "eval_loss": 0.2718713879585266, | |
| "eval_runtime": 408.121, | |
| "eval_samples_per_second": 20.34, | |
| "eval_steps_per_second": 5.087, | |
| "eval_wer": 0.25370000867528414, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 18.71, | |
| "learning_rate": 2.9502799552071662e-05, | |
| "loss": 1.016, | |
| "step": 29100 | |
| }, | |
| { | |
| "epoch": 18.78, | |
| "learning_rate": 2.9334826427771554e-05, | |
| "loss": 0.993, | |
| "step": 29200 | |
| }, | |
| { | |
| "epoch": 18.84, | |
| "learning_rate": 2.9166853303471442e-05, | |
| "loss": 0.9732, | |
| "step": 29300 | |
| }, | |
| { | |
| "epoch": 18.91, | |
| "learning_rate": 2.899888017917133e-05, | |
| "loss": 0.9605, | |
| "step": 29400 | |
| }, | |
| { | |
| "epoch": 18.97, | |
| "learning_rate": 2.8830907054871216e-05, | |
| "loss": 1.0157, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 18.97, | |
| "eval_loss": 0.26214534044265747, | |
| "eval_runtime": 408.8152, | |
| "eval_samples_per_second": 20.305, | |
| "eval_steps_per_second": 5.078, | |
| "eval_wer": 0.24486856944564933, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 19.04, | |
| "learning_rate": 2.8662933930571104e-05, | |
| "loss": 0.9566, | |
| "step": 29600 | |
| }, | |
| { | |
| "epoch": 19.1, | |
| "learning_rate": 2.8494960806270996e-05, | |
| "loss": 0.9643, | |
| "step": 29700 | |
| }, | |
| { | |
| "epoch": 19.16, | |
| "learning_rate": 2.8326987681970884e-05, | |
| "loss": 0.98, | |
| "step": 29800 | |
| }, | |
| { | |
| "epoch": 19.23, | |
| "learning_rate": 2.815901455767077e-05, | |
| "loss": 0.9953, | |
| "step": 29900 | |
| }, | |
| { | |
| "epoch": 19.29, | |
| "learning_rate": 2.7991041433370657e-05, | |
| "loss": 0.9572, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 19.29, | |
| "eval_loss": 0.2582398056983948, | |
| "eval_runtime": 408.3444, | |
| "eval_samples_per_second": 20.328, | |
| "eval_steps_per_second": 5.084, | |
| "eval_wer": 0.24273444955322288, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 19.36, | |
| "learning_rate": 2.7823068309070546e-05, | |
| "loss": 0.9839, | |
| "step": 30100 | |
| }, | |
| { | |
| "epoch": 19.42, | |
| "learning_rate": 2.7655095184770434e-05, | |
| "loss": 0.9709, | |
| "step": 30200 | |
| }, | |
| { | |
| "epoch": 19.49, | |
| "learning_rate": 2.748712206047032e-05, | |
| "loss": 0.9873, | |
| "step": 30300 | |
| }, | |
| { | |
| "epoch": 19.55, | |
| "learning_rate": 2.731914893617021e-05, | |
| "loss": 0.9743, | |
| "step": 30400 | |
| }, | |
| { | |
| "epoch": 19.61, | |
| "learning_rate": 2.71511758118701e-05, | |
| "loss": 0.9802, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 19.61, | |
| "eval_loss": 0.2707180380821228, | |
| "eval_runtime": 414.0735, | |
| "eval_samples_per_second": 20.047, | |
| "eval_steps_per_second": 5.014, | |
| "eval_wer": 0.2467597813828403, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 19.68, | |
| "learning_rate": 2.6983202687569987e-05, | |
| "loss": 0.9491, | |
| "step": 30600 | |
| }, | |
| { | |
| "epoch": 19.74, | |
| "learning_rate": 2.6815229563269872e-05, | |
| "loss": 0.9554, | |
| "step": 30700 | |
| }, | |
| { | |
| "epoch": 19.81, | |
| "learning_rate": 2.664725643896976e-05, | |
| "loss": 0.9593, | |
| "step": 30800 | |
| }, | |
| { | |
| "epoch": 19.87, | |
| "learning_rate": 2.6479283314669653e-05, | |
| "loss": 0.9385, | |
| "step": 30900 | |
| }, | |
| { | |
| "epoch": 19.94, | |
| "learning_rate": 2.631131019036954e-05, | |
| "loss": 0.9577, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 19.94, | |
| "eval_loss": 0.25633567571640015, | |
| "eval_runtime": 409.4741, | |
| "eval_samples_per_second": 20.272, | |
| "eval_steps_per_second": 5.07, | |
| "eval_wer": 0.23889997397414767, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 2.6143337066069426e-05, | |
| "loss": 0.9698, | |
| "step": 31100 | |
| }, | |
| { | |
| "epoch": 20.06, | |
| "learning_rate": 2.597704367301232e-05, | |
| "loss": 0.938, | |
| "step": 31200 | |
| }, | |
| { | |
| "epoch": 20.13, | |
| "learning_rate": 2.5809070548712203e-05, | |
| "loss": 0.941, | |
| "step": 31300 | |
| }, | |
| { | |
| "epoch": 20.19, | |
| "learning_rate": 2.564109742441209e-05, | |
| "loss": 0.921, | |
| "step": 31400 | |
| }, | |
| { | |
| "epoch": 20.26, | |
| "learning_rate": 2.547312430011198e-05, | |
| "loss": 0.9562, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 20.26, | |
| "eval_loss": 0.25917917490005493, | |
| "eval_runtime": 413.3121, | |
| "eval_samples_per_second": 20.084, | |
| "eval_steps_per_second": 5.023, | |
| "eval_wer": 0.23820595124490326, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 20.32, | |
| "learning_rate": 2.530515117581187e-05, | |
| "loss": 0.9329, | |
| "step": 31600 | |
| }, | |
| { | |
| "epoch": 20.39, | |
| "learning_rate": 2.5137178051511753e-05, | |
| "loss": 0.9464, | |
| "step": 31700 | |
| }, | |
| { | |
| "epoch": 20.45, | |
| "learning_rate": 2.496920492721164e-05, | |
| "loss": 0.9519, | |
| "step": 31800 | |
| }, | |
| { | |
| "epoch": 20.51, | |
| "learning_rate": 2.4801231802911533e-05, | |
| "loss": 0.973, | |
| "step": 31900 | |
| }, | |
| { | |
| "epoch": 20.58, | |
| "learning_rate": 2.4633258678611422e-05, | |
| "loss": 0.962, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 20.58, | |
| "eval_loss": 0.25392624735832214, | |
| "eval_runtime": 408.7718, | |
| "eval_samples_per_second": 20.307, | |
| "eval_steps_per_second": 5.079, | |
| "eval_wer": 0.2341285677105925, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 20.64, | |
| "learning_rate": 2.4465285554311307e-05, | |
| "loss": 0.9717, | |
| "step": 32100 | |
| }, | |
| { | |
| "epoch": 20.71, | |
| "learning_rate": 2.4297312430011195e-05, | |
| "loss": 0.9519, | |
| "step": 32200 | |
| }, | |
| { | |
| "epoch": 20.77, | |
| "learning_rate": 2.4129339305711083e-05, | |
| "loss": 0.9604, | |
| "step": 32300 | |
| }, | |
| { | |
| "epoch": 20.84, | |
| "learning_rate": 2.3961366181410975e-05, | |
| "loss": 0.9731, | |
| "step": 32400 | |
| }, | |
| { | |
| "epoch": 20.9, | |
| "learning_rate": 2.379339305711086e-05, | |
| "loss": 0.9541, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 20.9, | |
| "eval_loss": 0.25048619508743286, | |
| "eval_runtime": 409.2486, | |
| "eval_samples_per_second": 20.284, | |
| "eval_steps_per_second": 5.073, | |
| "eval_wer": 0.22878459269541077, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 20.96, | |
| "learning_rate": 2.362541993281075e-05, | |
| "loss": 0.9466, | |
| "step": 32600 | |
| }, | |
| { | |
| "epoch": 21.03, | |
| "learning_rate": 2.3457446808510637e-05, | |
| "loss": 0.9361, | |
| "step": 32700 | |
| }, | |
| { | |
| "epoch": 21.09, | |
| "learning_rate": 2.3289473684210525e-05, | |
| "loss": 0.9253, | |
| "step": 32800 | |
| }, | |
| { | |
| "epoch": 21.16, | |
| "learning_rate": 2.312150055991041e-05, | |
| "loss": 0.951, | |
| "step": 32900 | |
| }, | |
| { | |
| "epoch": 21.22, | |
| "learning_rate": 2.29535274356103e-05, | |
| "loss": 0.9587, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 21.22, | |
| "eval_loss": 0.24864241480827332, | |
| "eval_runtime": 410.1062, | |
| "eval_samples_per_second": 20.241, | |
| "eval_steps_per_second": 5.062, | |
| "eval_wer": 0.23022468985859287, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 21.29, | |
| "learning_rate": 2.278555431131019e-05, | |
| "loss": 0.9574, | |
| "step": 33100 | |
| }, | |
| { | |
| "epoch": 21.35, | |
| "learning_rate": 2.261758118701008e-05, | |
| "loss": 0.9261, | |
| "step": 33200 | |
| }, | |
| { | |
| "epoch": 21.41, | |
| "learning_rate": 2.2449608062709964e-05, | |
| "loss": 0.9216, | |
| "step": 33300 | |
| }, | |
| { | |
| "epoch": 21.48, | |
| "learning_rate": 2.2281634938409852e-05, | |
| "loss": 0.9108, | |
| "step": 33400 | |
| }, | |
| { | |
| "epoch": 21.54, | |
| "learning_rate": 2.211366181410974e-05, | |
| "loss": 0.9146, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 21.54, | |
| "eval_loss": 0.24606667459011078, | |
| "eval_runtime": 407.8433, | |
| "eval_samples_per_second": 20.353, | |
| "eval_steps_per_second": 5.09, | |
| "eval_wer": 0.22687603018998873, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 21.61, | |
| "learning_rate": 2.1945688689809632e-05, | |
| "loss": 0.9138, | |
| "step": 33600 | |
| }, | |
| { | |
| "epoch": 21.67, | |
| "learning_rate": 2.1777715565509517e-05, | |
| "loss": 0.9412, | |
| "step": 33700 | |
| }, | |
| { | |
| "epoch": 21.74, | |
| "learning_rate": 2.1609742441209405e-05, | |
| "loss": 0.9281, | |
| "step": 33800 | |
| }, | |
| { | |
| "epoch": 21.8, | |
| "learning_rate": 2.1441769316909294e-05, | |
| "loss": 0.9315, | |
| "step": 33900 | |
| }, | |
| { | |
| "epoch": 21.86, | |
| "learning_rate": 2.1273796192609182e-05, | |
| "loss": 0.9215, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 21.86, | |
| "eval_loss": 0.23870477080345154, | |
| "eval_runtime": 406.2395, | |
| "eval_samples_per_second": 20.434, | |
| "eval_steps_per_second": 5.11, | |
| "eval_wer": 0.22279864665567797, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 21.93, | |
| "learning_rate": 2.1105823068309067e-05, | |
| "loss": 0.924, | |
| "step": 34100 | |
| }, | |
| { | |
| "epoch": 21.99, | |
| "learning_rate": 2.0937849944008955e-05, | |
| "loss": 0.897, | |
| "step": 34200 | |
| }, | |
| { | |
| "epoch": 22.06, | |
| "learning_rate": 2.0769876819708847e-05, | |
| "loss": 0.8958, | |
| "step": 34300 | |
| }, | |
| { | |
| "epoch": 22.12, | |
| "learning_rate": 2.0601903695408732e-05, | |
| "loss": 0.8855, | |
| "step": 34400 | |
| }, | |
| { | |
| "epoch": 22.19, | |
| "learning_rate": 2.043393057110862e-05, | |
| "loss": 0.9105, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 22.19, | |
| "eval_loss": 0.24052534997463226, | |
| "eval_runtime": 408.199, | |
| "eval_samples_per_second": 20.336, | |
| "eval_steps_per_second": 5.086, | |
| "eval_wer": 0.22219137676758913, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 22.25, | |
| "learning_rate": 2.026595744680851e-05, | |
| "loss": 0.9043, | |
| "step": 34600 | |
| }, | |
| { | |
| "epoch": 22.32, | |
| "learning_rate": 2.0097984322508397e-05, | |
| "loss": 0.9173, | |
| "step": 34700 | |
| }, | |
| { | |
| "epoch": 22.38, | |
| "learning_rate": 1.9930011198208282e-05, | |
| "loss": 0.8962, | |
| "step": 34800 | |
| }, | |
| { | |
| "epoch": 22.44, | |
| "learning_rate": 1.9763717805151175e-05, | |
| "loss": 0.8795, | |
| "step": 34900 | |
| }, | |
| { | |
| "epoch": 22.51, | |
| "learning_rate": 1.9595744680851063e-05, | |
| "loss": 0.8949, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 22.51, | |
| "eval_loss": 0.2316313236951828, | |
| "eval_runtime": 410.5834, | |
| "eval_samples_per_second": 20.218, | |
| "eval_steps_per_second": 5.056, | |
| "eval_wer": 0.2190509239177583, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 22.57, | |
| "learning_rate": 1.9427771556550948e-05, | |
| "loss": 0.9183, | |
| "step": 35100 | |
| }, | |
| { | |
| "epoch": 22.64, | |
| "learning_rate": 1.9259798432250836e-05, | |
| "loss": 0.9169, | |
| "step": 35200 | |
| }, | |
| { | |
| "epoch": 22.7, | |
| "learning_rate": 1.9091825307950728e-05, | |
| "loss": 0.8924, | |
| "step": 35300 | |
| }, | |
| { | |
| "epoch": 22.77, | |
| "learning_rate": 1.8923852183650616e-05, | |
| "loss": 0.8782, | |
| "step": 35400 | |
| }, | |
| { | |
| "epoch": 22.83, | |
| "learning_rate": 1.87558790593505e-05, | |
| "loss": 0.9153, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 22.83, | |
| "eval_loss": 0.23579512536525726, | |
| "eval_runtime": 413.93, | |
| "eval_samples_per_second": 20.054, | |
| "eval_steps_per_second": 5.015, | |
| "eval_wer": 0.21799253925566062, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 22.89, | |
| "learning_rate": 1.858790593505039e-05, | |
| "loss": 0.8934, | |
| "step": 35600 | |
| }, | |
| { | |
| "epoch": 22.96, | |
| "learning_rate": 1.8419932810750278e-05, | |
| "loss": 0.9049, | |
| "step": 35700 | |
| }, | |
| { | |
| "epoch": 23.02, | |
| "learning_rate": 1.8251959686450166e-05, | |
| "loss": 0.8961, | |
| "step": 35800 | |
| }, | |
| { | |
| "epoch": 23.09, | |
| "learning_rate": 1.8083986562150055e-05, | |
| "loss": 0.881, | |
| "step": 35900 | |
| }, | |
| { | |
| "epoch": 23.15, | |
| "learning_rate": 1.7916013437849943e-05, | |
| "loss": 0.8907, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 23.15, | |
| "eval_loss": 0.23693053424358368, | |
| "eval_runtime": 413.4122, | |
| "eval_samples_per_second": 20.079, | |
| "eval_steps_per_second": 5.022, | |
| "eval_wer": 0.21677799947948295, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 23.22, | |
| "learning_rate": 1.774804031354983e-05, | |
| "loss": 0.8811, | |
| "step": 36100 | |
| }, | |
| { | |
| "epoch": 23.28, | |
| "learning_rate": 1.758006718924972e-05, | |
| "loss": 0.8909, | |
| "step": 36200 | |
| }, | |
| { | |
| "epoch": 23.34, | |
| "learning_rate": 1.7412094064949608e-05, | |
| "loss": 0.887, | |
| "step": 36300 | |
| }, | |
| { | |
| "epoch": 23.41, | |
| "learning_rate": 1.7244120940649493e-05, | |
| "loss": 0.8889, | |
| "step": 36400 | |
| }, | |
| { | |
| "epoch": 23.47, | |
| "learning_rate": 1.707614781634938e-05, | |
| "loss": 0.8973, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 23.47, | |
| "eval_loss": 0.23232492804527283, | |
| "eval_runtime": 410.3515, | |
| "eval_samples_per_second": 20.229, | |
| "eval_steps_per_second": 5.059, | |
| "eval_wer": 0.21202394378415892, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 23.54, | |
| "learning_rate": 1.690817469204927e-05, | |
| "loss": 0.8816, | |
| "step": 36600 | |
| }, | |
| { | |
| "epoch": 23.6, | |
| "learning_rate": 1.6740201567749158e-05, | |
| "loss": 0.926, | |
| "step": 36700 | |
| }, | |
| { | |
| "epoch": 23.67, | |
| "learning_rate": 1.6572228443449046e-05, | |
| "loss": 0.8609, | |
| "step": 36800 | |
| }, | |
| { | |
| "epoch": 23.73, | |
| "learning_rate": 1.6404255319148935e-05, | |
| "loss": 0.9076, | |
| "step": 36900 | |
| }, | |
| { | |
| "epoch": 23.79, | |
| "learning_rate": 1.6236282194848823e-05, | |
| "loss": 0.8878, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 23.79, | |
| "eval_loss": 0.2293161004781723, | |
| "eval_runtime": 410.3781, | |
| "eval_samples_per_second": 20.228, | |
| "eval_steps_per_second": 5.059, | |
| "eval_wer": 0.21041034093866573, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 23.86, | |
| "learning_rate": 1.6069988801791712e-05, | |
| "loss": 0.905, | |
| "step": 37100 | |
| }, | |
| { | |
| "epoch": 23.92, | |
| "learning_rate": 1.59020156774916e-05, | |
| "loss": 0.8886, | |
| "step": 37200 | |
| }, | |
| { | |
| "epoch": 23.99, | |
| "learning_rate": 1.573404255319149e-05, | |
| "loss": 0.9052, | |
| "step": 37300 | |
| }, | |
| { | |
| "epoch": 24.05, | |
| "learning_rate": 1.5566069428891377e-05, | |
| "loss": 0.8697, | |
| "step": 37400 | |
| }, | |
| { | |
| "epoch": 24.12, | |
| "learning_rate": 1.5398096304591266e-05, | |
| "loss": 0.8818, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 24.12, | |
| "eval_loss": 0.23020902276039124, | |
| "eval_runtime": 410.0065, | |
| "eval_samples_per_second": 20.246, | |
| "eval_steps_per_second": 5.063, | |
| "eval_wer": 0.21318643185564326, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 24.18, | |
| "learning_rate": 1.5230123180291152e-05, | |
| "loss": 0.9025, | |
| "step": 37600 | |
| }, | |
| { | |
| "epoch": 24.24, | |
| "learning_rate": 1.506215005599104e-05, | |
| "loss": 0.8684, | |
| "step": 37700 | |
| }, | |
| { | |
| "epoch": 24.31, | |
| "learning_rate": 1.4894176931690927e-05, | |
| "loss": 0.8614, | |
| "step": 37800 | |
| }, | |
| { | |
| "epoch": 24.37, | |
| "learning_rate": 1.4726203807390817e-05, | |
| "loss": 0.8674, | |
| "step": 37900 | |
| }, | |
| { | |
| "epoch": 24.44, | |
| "learning_rate": 1.4558230683090704e-05, | |
| "loss": 0.8919, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 24.44, | |
| "eval_loss": 0.2261827141046524, | |
| "eval_runtime": 408.9855, | |
| "eval_samples_per_second": 20.297, | |
| "eval_steps_per_second": 5.076, | |
| "eval_wer": 0.20831092218270147, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 24.5, | |
| "learning_rate": 1.4390257558790594e-05, | |
| "loss": 0.8693, | |
| "step": 38100 | |
| }, | |
| { | |
| "epoch": 24.57, | |
| "learning_rate": 1.4223964165733481e-05, | |
| "loss": 0.8611, | |
| "step": 38200 | |
| }, | |
| { | |
| "epoch": 24.63, | |
| "learning_rate": 1.405599104143337e-05, | |
| "loss": 0.8611, | |
| "step": 38300 | |
| }, | |
| { | |
| "epoch": 24.69, | |
| "learning_rate": 1.3888017917133256e-05, | |
| "loss": 0.8831, | |
| "step": 38400 | |
| }, | |
| { | |
| "epoch": 24.76, | |
| "learning_rate": 1.3720044792833146e-05, | |
| "loss": 0.8473, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 24.76, | |
| "eval_loss": 0.22569720447063446, | |
| "eval_runtime": 408.2862, | |
| "eval_samples_per_second": 20.331, | |
| "eval_steps_per_second": 5.085, | |
| "eval_wer": 0.20397328012492408, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 24.82, | |
| "learning_rate": 1.3552071668533033e-05, | |
| "loss": 0.8417, | |
| "step": 38600 | |
| }, | |
| { | |
| "epoch": 24.89, | |
| "learning_rate": 1.3384098544232921e-05, | |
| "loss": 0.8559, | |
| "step": 38700 | |
| }, | |
| { | |
| "epoch": 24.95, | |
| "learning_rate": 1.321612541993281e-05, | |
| "loss": 0.8865, | |
| "step": 38800 | |
| }, | |
| { | |
| "epoch": 25.02, | |
| "learning_rate": 1.3048152295632698e-05, | |
| "loss": 0.8492, | |
| "step": 38900 | |
| }, | |
| { | |
| "epoch": 25.08, | |
| "learning_rate": 1.2880179171332585e-05, | |
| "loss": 0.8516, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 25.08, | |
| "eval_loss": 0.22461159527301788, | |
| "eval_runtime": 409.5188, | |
| "eval_samples_per_second": 20.27, | |
| "eval_steps_per_second": 5.069, | |
| "eval_wer": 0.2030710505769064, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 25.14, | |
| "learning_rate": 1.2712206047032475e-05, | |
| "loss": 0.8566, | |
| "step": 39100 | |
| }, | |
| { | |
| "epoch": 25.21, | |
| "learning_rate": 1.2544232922732362e-05, | |
| "loss": 0.8576, | |
| "step": 39200 | |
| }, | |
| { | |
| "epoch": 25.27, | |
| "learning_rate": 1.237625979843225e-05, | |
| "loss": 0.8687, | |
| "step": 39300 | |
| }, | |
| { | |
| "epoch": 25.34, | |
| "learning_rate": 1.2208286674132138e-05, | |
| "loss": 0.8481, | |
| "step": 39400 | |
| }, | |
| { | |
| "epoch": 25.4, | |
| "learning_rate": 1.2040313549832027e-05, | |
| "loss": 0.8451, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 25.4, | |
| "eval_loss": 0.21983762085437775, | |
| "eval_runtime": 409.8341, | |
| "eval_samples_per_second": 20.255, | |
| "eval_steps_per_second": 5.065, | |
| "eval_wer": 0.19996529886353778, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 25.47, | |
| "learning_rate": 1.1872340425531913e-05, | |
| "loss": 0.838, | |
| "step": 39600 | |
| }, | |
| { | |
| "epoch": 25.53, | |
| "learning_rate": 1.1704367301231803e-05, | |
| "loss": 0.8601, | |
| "step": 39700 | |
| }, | |
| { | |
| "epoch": 25.59, | |
| "learning_rate": 1.153639417693169e-05, | |
| "loss": 0.849, | |
| "step": 39800 | |
| }, | |
| { | |
| "epoch": 25.66, | |
| "learning_rate": 1.1368421052631578e-05, | |
| "loss": 0.8263, | |
| "step": 39900 | |
| }, | |
| { | |
| "epoch": 25.72, | |
| "learning_rate": 1.1200447928331467e-05, | |
| "loss": 0.8288, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 25.72, | |
| "eval_loss": 0.21986670792102814, | |
| "eval_runtime": 409.1679, | |
| "eval_samples_per_second": 20.288, | |
| "eval_steps_per_second": 5.074, | |
| "eval_wer": 0.19902836817905786, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 25.79, | |
| "learning_rate": 1.1032474804031355e-05, | |
| "loss": 0.8439, | |
| "step": 40100 | |
| }, | |
| { | |
| "epoch": 25.85, | |
| "learning_rate": 1.0864501679731242e-05, | |
| "loss": 0.833, | |
| "step": 40200 | |
| }, | |
| { | |
| "epoch": 25.92, | |
| "learning_rate": 1.0696528555431132e-05, | |
| "loss": 0.8356, | |
| "step": 40300 | |
| }, | |
| { | |
| "epoch": 25.98, | |
| "learning_rate": 1.0528555431131018e-05, | |
| "loss": 0.8326, | |
| "step": 40400 | |
| }, | |
| { | |
| "epoch": 26.05, | |
| "learning_rate": 1.0360582306830907e-05, | |
| "loss": 0.8465, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 26.05, | |
| "eval_loss": 0.2165078967809677, | |
| "eval_runtime": 408.4852, | |
| "eval_samples_per_second": 20.321, | |
| "eval_steps_per_second": 5.082, | |
| "eval_wer": 0.19718920794656025, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 26.11, | |
| "learning_rate": 1.0192609182530793e-05, | |
| "loss": 0.8194, | |
| "step": 40600 | |
| }, | |
| { | |
| "epoch": 26.17, | |
| "learning_rate": 1.0024636058230683e-05, | |
| "loss": 0.8219, | |
| "step": 40700 | |
| }, | |
| { | |
| "epoch": 26.24, | |
| "learning_rate": 9.85666293393057e-06, | |
| "loss": 0.852, | |
| "step": 40800 | |
| }, | |
| { | |
| "epoch": 26.3, | |
| "learning_rate": 9.688689809630457e-06, | |
| "loss": 0.8342, | |
| "step": 40900 | |
| }, | |
| { | |
| "epoch": 26.37, | |
| "learning_rate": 9.520716685330347e-06, | |
| "loss": 0.8305, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 26.37, | |
| "eval_loss": 0.21275557577610016, | |
| "eval_runtime": 416.3044, | |
| "eval_samples_per_second": 19.94, | |
| "eval_steps_per_second": 4.987, | |
| "eval_wer": 0.19571440964691594, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 26.43, | |
| "learning_rate": 9.352743561030235e-06, | |
| "loss": 0.84, | |
| "step": 41100 | |
| }, | |
| { | |
| "epoch": 26.5, | |
| "learning_rate": 9.184770436730122e-06, | |
| "loss": 0.8359, | |
| "step": 41200 | |
| }, | |
| { | |
| "epoch": 26.56, | |
| "learning_rate": 9.01679731243001e-06, | |
| "loss": 0.8126, | |
| "step": 41300 | |
| }, | |
| { | |
| "epoch": 26.62, | |
| "learning_rate": 8.848824188129898e-06, | |
| "loss": 0.8445, | |
| "step": 41400 | |
| }, | |
| { | |
| "epoch": 26.69, | |
| "learning_rate": 8.680851063829787e-06, | |
| "loss": 0.8202, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 26.69, | |
| "eval_loss": 0.2126561850309372, | |
| "eval_runtime": 409.831, | |
| "eval_samples_per_second": 20.255, | |
| "eval_steps_per_second": 5.066, | |
| "eval_wer": 0.19366704259564502, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 26.75, | |
| "learning_rate": 8.512877939529675e-06, | |
| "loss": 0.8424, | |
| "step": 41600 | |
| }, | |
| { | |
| "epoch": 26.82, | |
| "learning_rate": 8.344904815229563e-06, | |
| "loss": 0.8037, | |
| "step": 41700 | |
| }, | |
| { | |
| "epoch": 26.88, | |
| "learning_rate": 8.17693169092945e-06, | |
| "loss": 0.8075, | |
| "step": 41800 | |
| }, | |
| { | |
| "epoch": 26.95, | |
| "learning_rate": 8.008958566629338e-06, | |
| "loss": 0.8062, | |
| "step": 41900 | |
| }, | |
| { | |
| "epoch": 27.01, | |
| "learning_rate": 7.840985442329227e-06, | |
| "loss": 0.8223, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 27.01, | |
| "eval_loss": 0.21001781523227692, | |
| "eval_runtime": 410.3414, | |
| "eval_samples_per_second": 20.229, | |
| "eval_steps_per_second": 5.059, | |
| "eval_wer": 0.19335473236748504, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 27.07, | |
| "learning_rate": 7.673012318029115e-06, | |
| "loss": 0.8233, | |
| "step": 42100 | |
| }, | |
| { | |
| "epoch": 27.14, | |
| "learning_rate": 7.5050391937290035e-06, | |
| "loss": 0.8278, | |
| "step": 42200 | |
| }, | |
| { | |
| "epoch": 27.2, | |
| "learning_rate": 7.337066069428891e-06, | |
| "loss": 0.8212, | |
| "step": 42300 | |
| }, | |
| { | |
| "epoch": 27.27, | |
| "learning_rate": 7.17077267637178e-06, | |
| "loss": 0.8306, | |
| "step": 42400 | |
| }, | |
| { | |
| "epoch": 27.33, | |
| "learning_rate": 7.004479283314669e-06, | |
| "loss": 0.8322, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 27.33, | |
| "eval_loss": 0.20762594044208527, | |
| "eval_runtime": 410.9869, | |
| "eval_samples_per_second": 20.198, | |
| "eval_steps_per_second": 5.051, | |
| "eval_wer": 0.19054394031404528, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 27.4, | |
| "learning_rate": 6.836506159014558e-06, | |
| "loss": 0.7888, | |
| "step": 42600 | |
| }, | |
| { | |
| "epoch": 27.46, | |
| "learning_rate": 6.668533034714445e-06, | |
| "loss": 0.799, | |
| "step": 42700 | |
| }, | |
| { | |
| "epoch": 27.52, | |
| "learning_rate": 6.5005599104143335e-06, | |
| "loss": 0.7908, | |
| "step": 42800 | |
| }, | |
| { | |
| "epoch": 27.59, | |
| "learning_rate": 6.332586786114222e-06, | |
| "loss": 0.8117, | |
| "step": 42900 | |
| }, | |
| { | |
| "epoch": 27.65, | |
| "learning_rate": 6.164613661814109e-06, | |
| "loss": 0.8139, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 27.65, | |
| "eval_loss": 0.2053818255662918, | |
| "eval_runtime": 411.3503, | |
| "eval_samples_per_second": 20.18, | |
| "eval_steps_per_second": 5.047, | |
| "eval_wer": 0.1880454584887655, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 27.72, | |
| "learning_rate": 5.996640537513998e-06, | |
| "loss": 0.8055, | |
| "step": 43100 | |
| }, | |
| { | |
| "epoch": 27.78, | |
| "learning_rate": 5.828667413213886e-06, | |
| "loss": 0.8035, | |
| "step": 43200 | |
| }, | |
| { | |
| "epoch": 27.85, | |
| "learning_rate": 5.6606942889137735e-06, | |
| "loss": 0.8202, | |
| "step": 43300 | |
| }, | |
| { | |
| "epoch": 27.91, | |
| "learning_rate": 5.492721164613661e-06, | |
| "loss": 0.8267, | |
| "step": 43400 | |
| }, | |
| { | |
| "epoch": 27.97, | |
| "learning_rate": 5.3247480403135485e-06, | |
| "loss": 0.8299, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 27.97, | |
| "eval_loss": 0.2026369571685791, | |
| "eval_runtime": 413.8334, | |
| "eval_samples_per_second": 20.059, | |
| "eval_steps_per_second": 5.017, | |
| "eval_wer": 0.18681356814435673, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 28.04, | |
| "learning_rate": 5.156774916013437e-06, | |
| "loss": 0.7985, | |
| "step": 43600 | |
| }, | |
| { | |
| "epoch": 28.1, | |
| "learning_rate": 4.988801791713325e-06, | |
| "loss": 0.7826, | |
| "step": 43700 | |
| }, | |
| { | |
| "epoch": 28.17, | |
| "learning_rate": 4.820828667413213e-06, | |
| "loss": 0.8098, | |
| "step": 43800 | |
| }, | |
| { | |
| "epoch": 28.23, | |
| "learning_rate": 4.652855543113102e-06, | |
| "loss": 0.7905, | |
| "step": 43900 | |
| }, | |
| { | |
| "epoch": 28.3, | |
| "learning_rate": 4.484882418812989e-06, | |
| "loss": 0.7937, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 28.3, | |
| "eval_loss": 0.20451003313064575, | |
| "eval_runtime": 411.9784, | |
| "eval_samples_per_second": 20.149, | |
| "eval_steps_per_second": 5.039, | |
| "eval_wer": 0.18721263121367224, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 28.36, | |
| "learning_rate": 4.316909294512878e-06, | |
| "loss": 0.816, | |
| "step": 44100 | |
| }, | |
| { | |
| "epoch": 28.42, | |
| "learning_rate": 4.148936170212765e-06, | |
| "loss": 0.7819, | |
| "step": 44200 | |
| }, | |
| { | |
| "epoch": 28.49, | |
| "learning_rate": 3.980963045912654e-06, | |
| "loss": 0.7981, | |
| "step": 44300 | |
| }, | |
| { | |
| "epoch": 28.55, | |
| "learning_rate": 3.8129899216125415e-06, | |
| "loss": 0.7832, | |
| "step": 44400 | |
| }, | |
| { | |
| "epoch": 28.62, | |
| "learning_rate": 3.6450167973124295e-06, | |
| "loss": 0.7972, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 28.62, | |
| "eval_loss": 0.20249634981155396, | |
| "eval_runtime": 416.0753, | |
| "eval_samples_per_second": 19.951, | |
| "eval_steps_per_second": 4.989, | |
| "eval_wer": 0.18613689598334346, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 28.68, | |
| "learning_rate": 3.4770436730123174e-06, | |
| "loss": 0.7895, | |
| "step": 44600 | |
| }, | |
| { | |
| "epoch": 28.75, | |
| "learning_rate": 3.3090705487122057e-06, | |
| "loss": 0.7918, | |
| "step": 44700 | |
| }, | |
| { | |
| "epoch": 28.81, | |
| "learning_rate": 3.1410974244120937e-06, | |
| "loss": 0.7907, | |
| "step": 44800 | |
| }, | |
| { | |
| "epoch": 28.87, | |
| "learning_rate": 2.9731243001119816e-06, | |
| "loss": 0.7851, | |
| "step": 44900 | |
| }, | |
| { | |
| "epoch": 28.94, | |
| "learning_rate": 2.80515117581187e-06, | |
| "loss": 0.809, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 28.94, | |
| "eval_loss": 0.20257468521595, | |
| "eval_runtime": 412.9649, | |
| "eval_samples_per_second": 20.101, | |
| "eval_steps_per_second": 5.027, | |
| "eval_wer": 0.18575518348225906, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "learning_rate": 2.6388577827547587e-06, | |
| "loss": 0.7747, | |
| "step": 45100 | |
| }, | |
| { | |
| "epoch": 29.07, | |
| "learning_rate": 2.470884658454647e-06, | |
| "loss": 0.7852, | |
| "step": 45200 | |
| }, | |
| { | |
| "epoch": 29.13, | |
| "learning_rate": 2.302911534154535e-06, | |
| "loss": 0.7897, | |
| "step": 45300 | |
| }, | |
| { | |
| "epoch": 29.2, | |
| "learning_rate": 2.134938409854423e-06, | |
| "loss": 0.7691, | |
| "step": 45400 | |
| }, | |
| { | |
| "epoch": 29.26, | |
| "learning_rate": 1.966965285554311e-06, | |
| "loss": 0.813, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 29.26, | |
| "eval_loss": 0.20126082003116608, | |
| "eval_runtime": 410.0758, | |
| "eval_samples_per_second": 20.243, | |
| "eval_steps_per_second": 5.062, | |
| "eval_wer": 0.183846620976837, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 29.32, | |
| "learning_rate": 1.7989921612541991e-06, | |
| "loss": 0.7961, | |
| "step": 45600 | |
| }, | |
| { | |
| "epoch": 29.39, | |
| "learning_rate": 1.6326987681970883e-06, | |
| "loss": 0.7824, | |
| "step": 45700 | |
| }, | |
| { | |
| "epoch": 29.45, | |
| "learning_rate": 1.4647256438969762e-06, | |
| "loss": 0.7992, | |
| "step": 45800 | |
| }, | |
| { | |
| "epoch": 29.52, | |
| "learning_rate": 1.2967525195968644e-06, | |
| "loss": 0.7726, | |
| "step": 45900 | |
| }, | |
| { | |
| "epoch": 29.58, | |
| "learning_rate": 1.1287793952967525e-06, | |
| "loss": 0.7718, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 29.58, | |
| "eval_loss": 0.20103685557842255, | |
| "eval_runtime": 412.3355, | |
| "eval_samples_per_second": 20.132, | |
| "eval_steps_per_second": 5.035, | |
| "eval_wer": 0.18372516699921923, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 29.65, | |
| "learning_rate": 9.608062709966404e-07, | |
| "loss": 0.768, | |
| "step": 46100 | |
| }, | |
| { | |
| "epoch": 29.71, | |
| "learning_rate": 7.928331466965285e-07, | |
| "loss": 0.758, | |
| "step": 46200 | |
| }, | |
| { | |
| "epoch": 29.77, | |
| "learning_rate": 6.248600223964165e-07, | |
| "loss": 0.7826, | |
| "step": 46300 | |
| }, | |
| { | |
| "epoch": 29.84, | |
| "learning_rate": 4.5688689809630456e-07, | |
| "loss": 0.8091, | |
| "step": 46400 | |
| }, | |
| { | |
| "epoch": 29.9, | |
| "learning_rate": 2.889137737961926e-07, | |
| "loss": 0.7929, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 29.9, | |
| "eval_loss": 0.20083273947238922, | |
| "eval_runtime": 411.6027, | |
| "eval_samples_per_second": 20.168, | |
| "eval_steps_per_second": 5.044, | |
| "eval_wer": 0.18398542552268587, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 29.97, | |
| "learning_rate": 1.2094064949608062e-07, | |
| "loss": 0.795, | |
| "step": 46600 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "step": 46650, | |
| "total_flos": 2.389628176512902e+20, | |
| "train_loss": 1.120528796197006, | |
| "train_runtime": 125405.8596, | |
| "train_samples_per_second": 5.951, | |
| "train_steps_per_second": 0.372 | |
| } | |
| ], | |
| "max_steps": 46650, | |
| "num_train_epochs": 30, | |
| "total_flos": 2.389628176512902e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |