| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 60.0, | |
| "global_step": 9960, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.8e-06, | |
| "loss": 27.2555, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 9.799999999999998e-06, | |
| "loss": 12.0667, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 1.4799999999999999e-05, | |
| "loss": 6.8407, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 1.98e-05, | |
| "loss": 5.175, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "eval_loss": 4.690936088562012, | |
| "eval_runtime": 87.6676, | |
| "eval_samples_per_second": 25.289, | |
| "eval_steps_per_second": 1.586, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 2.4799999999999996e-05, | |
| "loss": 4.1143, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 2.9799999999999996e-05, | |
| "loss": 3.5613, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 3.48e-05, | |
| "loss": 3.4854, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "learning_rate": 3.979999999999999e-05, | |
| "loss": 3.3785, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "eval_loss": 3.308009147644043, | |
| "eval_runtime": 87.3792, | |
| "eval_samples_per_second": 25.372, | |
| "eval_steps_per_second": 1.591, | |
| "eval_wer": 1.0, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.42, | |
| "learning_rate": 4.48e-05, | |
| "loss": 3.2155, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 4.98e-05, | |
| "loss": 3.1587, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.63, | |
| "learning_rate": 5.48e-05, | |
| "loss": 3.0407, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 7.23, | |
| "learning_rate": 5.98e-05, | |
| "loss": 2.6964, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 7.23, | |
| "eval_loss": 2.0650923252105713, | |
| "eval_runtime": 86.9803, | |
| "eval_samples_per_second": 25.489, | |
| "eval_steps_per_second": 1.598, | |
| "eval_wer": 1.105476923076923, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "learning_rate": 6.479999999999999e-05, | |
| "loss": 2.0851, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 8.43, | |
| "learning_rate": 6.979999999999999e-05, | |
| "loss": 1.6595, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 7.479999999999999e-05, | |
| "loss": 1.4478, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 9.64, | |
| "learning_rate": 7.414893617021276e-05, | |
| "loss": 1.3008, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 9.64, | |
| "eval_loss": 0.5845397114753723, | |
| "eval_runtime": 86.5846, | |
| "eval_samples_per_second": 25.605, | |
| "eval_steps_per_second": 1.605, | |
| "eval_wer": 0.6206769230769231, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 10.24, | |
| "learning_rate": 7.326241134751773e-05, | |
| "loss": 1.2563, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 10.84, | |
| "learning_rate": 7.237588652482268e-05, | |
| "loss": 1.1552, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 11.45, | |
| "learning_rate": 7.148936170212766e-05, | |
| "loss": 1.1357, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 7.060283687943261e-05, | |
| "loss": 1.1185, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "eval_loss": 0.4195278584957123, | |
| "eval_runtime": 87.4194, | |
| "eval_samples_per_second": 25.361, | |
| "eval_steps_per_second": 1.59, | |
| "eval_wer": 0.4193230769230769, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 12.65, | |
| "learning_rate": 6.971631205673758e-05, | |
| "loss": 1.0362, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 13.25, | |
| "learning_rate": 6.882978723404255e-05, | |
| "loss": 1.0465, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 13.86, | |
| "learning_rate": 6.79432624113475e-05, | |
| "loss": 0.984, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 14.46, | |
| "learning_rate": 6.705673758865247e-05, | |
| "loss": 1.0252, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 14.46, | |
| "eval_loss": 0.3823728859424591, | |
| "eval_runtime": 87.0092, | |
| "eval_samples_per_second": 25.48, | |
| "eval_steps_per_second": 1.598, | |
| "eval_wer": 0.35704615384615385, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 15.06, | |
| "learning_rate": 6.617021276595744e-05, | |
| "loss": 0.9813, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 15.66, | |
| "learning_rate": 6.52836879432624e-05, | |
| "loss": 0.9642, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 16.27, | |
| "learning_rate": 6.439716312056737e-05, | |
| "loss": 0.9494, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 16.87, | |
| "learning_rate": 6.351063829787234e-05, | |
| "loss": 0.935, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 16.87, | |
| "eval_loss": 0.36926984786987305, | |
| "eval_runtime": 88.3011, | |
| "eval_samples_per_second": 25.107, | |
| "eval_steps_per_second": 1.574, | |
| "eval_wer": 0.3462153846153846, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 17.47, | |
| "learning_rate": 6.26241134751773e-05, | |
| "loss": 0.9282, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 18.07, | |
| "learning_rate": 6.173758865248226e-05, | |
| "loss": 0.8868, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 18.67, | |
| "learning_rate": 6.0851063829787224e-05, | |
| "loss": 0.8898, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 19.28, | |
| "learning_rate": 5.99645390070922e-05, | |
| "loss": 0.8818, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 19.28, | |
| "eval_loss": 0.3587293326854706, | |
| "eval_runtime": 87.4375, | |
| "eval_samples_per_second": 25.355, | |
| "eval_steps_per_second": 1.59, | |
| "eval_wer": 0.3318153846153846, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 19.88, | |
| "learning_rate": 5.907801418439716e-05, | |
| "loss": 0.8633, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 20.48, | |
| "learning_rate": 5.819148936170212e-05, | |
| "loss": 0.8942, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 21.08, | |
| "learning_rate": 5.730496453900709e-05, | |
| "loss": 0.856, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 21.69, | |
| "learning_rate": 5.641843971631205e-05, | |
| "loss": 0.8534, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 21.69, | |
| "eval_loss": 0.3420199751853943, | |
| "eval_runtime": 88.0755, | |
| "eval_samples_per_second": 25.172, | |
| "eval_steps_per_second": 1.578, | |
| "eval_wer": 0.31803076923076923, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 22.29, | |
| "learning_rate": 5.5531914893617016e-05, | |
| "loss": 0.8509, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 22.89, | |
| "learning_rate": 5.4645390070921984e-05, | |
| "loss": 0.8185, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 23.49, | |
| "learning_rate": 5.375886524822695e-05, | |
| "loss": 0.8319, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 24.1, | |
| "learning_rate": 5.287234042553191e-05, | |
| "loss": 0.8137, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 24.1, | |
| "eval_loss": 0.34257009625434875, | |
| "eval_runtime": 89.8454, | |
| "eval_samples_per_second": 24.676, | |
| "eval_steps_per_second": 1.547, | |
| "eval_wer": 0.31298461538461536, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 24.7, | |
| "learning_rate": 5.198581560283687e-05, | |
| "loss": 0.8089, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 25.3, | |
| "learning_rate": 5.110815602836879e-05, | |
| "loss": 0.8291, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 25.9, | |
| "learning_rate": 5.0221631205673754e-05, | |
| "loss": 0.7874, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 26.51, | |
| "learning_rate": 4.9335106382978715e-05, | |
| "loss": 0.7968, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 26.51, | |
| "eval_loss": 0.334939181804657, | |
| "eval_runtime": 85.418, | |
| "eval_samples_per_second": 25.955, | |
| "eval_steps_per_second": 1.627, | |
| "eval_wer": 0.31015384615384617, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 27.11, | |
| "learning_rate": 4.844858156028368e-05, | |
| "loss": 0.7791, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 27.71, | |
| "learning_rate": 4.756205673758865e-05, | |
| "loss": 0.7802, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 28.31, | |
| "learning_rate": 4.667553191489362e-05, | |
| "loss": 0.7639, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 28.92, | |
| "learning_rate": 4.579787234042553e-05, | |
| "loss": 0.7558, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 28.92, | |
| "eval_loss": 0.3215944170951843, | |
| "eval_runtime": 85.8433, | |
| "eval_samples_per_second": 25.826, | |
| "eval_steps_per_second": 1.619, | |
| "eval_wer": 0.3019076923076923, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 29.52, | |
| "learning_rate": 4.49113475177305e-05, | |
| "loss": 0.8011, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 30.12, | |
| "learning_rate": 4.402482269503546e-05, | |
| "loss": 0.7381, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 30.72, | |
| "learning_rate": 4.313829787234042e-05, | |
| "loss": 0.7412, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 31.33, | |
| "learning_rate": 4.225177304964539e-05, | |
| "loss": 0.7313, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 31.33, | |
| "eval_loss": 0.3450605571269989, | |
| "eval_runtime": 86.5107, | |
| "eval_samples_per_second": 25.627, | |
| "eval_steps_per_second": 1.607, | |
| "eval_wer": 0.30596923076923077, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 31.93, | |
| "learning_rate": 4.136524822695035e-05, | |
| "loss": 0.7421, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 32.53, | |
| "learning_rate": 4.047872340425531e-05, | |
| "loss": 0.7639, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 33.13, | |
| "learning_rate": 3.9592198581560283e-05, | |
| "loss": 0.7198, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 33.73, | |
| "learning_rate": 3.8705673758865244e-05, | |
| "loss": 0.7358, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 33.73, | |
| "eval_loss": 0.32715940475463867, | |
| "eval_runtime": 89.1066, | |
| "eval_samples_per_second": 24.88, | |
| "eval_steps_per_second": 1.56, | |
| "eval_wer": 0.29673846153846156, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 34.34, | |
| "learning_rate": 3.781914893617021e-05, | |
| "loss": 0.7053, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 34.94, | |
| "learning_rate": 3.693262411347517e-05, | |
| "loss": 0.7084, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 35.54, | |
| "learning_rate": 3.604609929078014e-05, | |
| "loss": 0.7029, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 36.14, | |
| "learning_rate": 3.51595744680851e-05, | |
| "loss": 0.718, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 36.14, | |
| "eval_loss": 0.3315288722515106, | |
| "eval_runtime": 85.6491, | |
| "eval_samples_per_second": 25.885, | |
| "eval_steps_per_second": 1.623, | |
| "eval_wer": 0.2882461538461538, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 36.75, | |
| "learning_rate": 3.427304964539007e-05, | |
| "loss": 0.7136, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 37.35, | |
| "learning_rate": 3.338652482269503e-05, | |
| "loss": 0.6807, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 37.95, | |
| "learning_rate": 3.25e-05, | |
| "loss": 0.6968, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 38.55, | |
| "learning_rate": 3.1613475177304965e-05, | |
| "loss": 0.6991, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 38.55, | |
| "eval_loss": 0.3299025297164917, | |
| "eval_runtime": 91.3334, | |
| "eval_samples_per_second": 24.274, | |
| "eval_steps_per_second": 1.522, | |
| "eval_wer": 0.28295384615384617, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 39.16, | |
| "learning_rate": 3.0726950354609926e-05, | |
| "loss": 0.6666, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 39.76, | |
| "learning_rate": 2.9840425531914893e-05, | |
| "loss": 0.669, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 40.36, | |
| "learning_rate": 2.8953900709219857e-05, | |
| "loss": 0.6367, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 40.96, | |
| "learning_rate": 2.8067375886524818e-05, | |
| "loss": 0.6529, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 40.96, | |
| "eval_loss": 0.3140033185482025, | |
| "eval_runtime": 86.1836, | |
| "eval_samples_per_second": 25.724, | |
| "eval_steps_per_second": 1.613, | |
| "eval_wer": 0.2835692307692308, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 41.57, | |
| "learning_rate": 2.7180851063829786e-05, | |
| "loss": 0.6539, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 42.17, | |
| "learning_rate": 2.629432624113475e-05, | |
| "loss": 0.6421, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 42.77, | |
| "learning_rate": 2.5407801418439714e-05, | |
| "loss": 0.6533, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 43.37, | |
| "learning_rate": 2.4521276595744675e-05, | |
| "loss": 0.6225, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 43.37, | |
| "eval_loss": 0.31283286213874817, | |
| "eval_runtime": 89.5905, | |
| "eval_samples_per_second": 24.746, | |
| "eval_steps_per_second": 1.552, | |
| "eval_wer": 0.27507692307692305, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 43.98, | |
| "learning_rate": 2.3634751773049643e-05, | |
| "loss": 0.6544, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 44.58, | |
| "learning_rate": 2.2748226950354607e-05, | |
| "loss": 0.626, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 45.18, | |
| "learning_rate": 2.186170212765957e-05, | |
| "loss": 0.6182, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 45.78, | |
| "learning_rate": 2.097517730496454e-05, | |
| "loss": 0.633, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 45.78, | |
| "eval_loss": 0.3211357891559601, | |
| "eval_runtime": 86.4578, | |
| "eval_samples_per_second": 25.643, | |
| "eval_steps_per_second": 1.608, | |
| "eval_wer": 0.2774153846153846, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 46.39, | |
| "learning_rate": 2.0097517730496455e-05, | |
| "loss": 0.6123, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 46.99, | |
| "learning_rate": 1.9210992907801416e-05, | |
| "loss": 0.6536, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 47.59, | |
| "learning_rate": 1.832446808510638e-05, | |
| "loss": 0.614, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 48.19, | |
| "learning_rate": 1.7437943262411345e-05, | |
| "loss": 0.5876, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 48.19, | |
| "eval_loss": 0.3162022531032562, | |
| "eval_runtime": 87.3839, | |
| "eval_samples_per_second": 25.371, | |
| "eval_steps_per_second": 1.591, | |
| "eval_wer": 0.27643076923076926, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 48.8, | |
| "learning_rate": 1.656028368794326e-05, | |
| "loss": 0.6021, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 49.4, | |
| "learning_rate": 1.5673758865248225e-05, | |
| "loss": 0.5854, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 1.4787234042553191e-05, | |
| "loss": 0.6204, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 50.6, | |
| "learning_rate": 1.3900709219858154e-05, | |
| "loss": 0.588, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 50.6, | |
| "eval_loss": 0.30821505188941956, | |
| "eval_runtime": 91.283, | |
| "eval_samples_per_second": 24.287, | |
| "eval_steps_per_second": 1.523, | |
| "eval_wer": 0.27224615384615386, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 51.2, | |
| "learning_rate": 1.301418439716312e-05, | |
| "loss": 0.5872, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 51.81, | |
| "learning_rate": 1.2127659574468086e-05, | |
| "loss": 0.5996, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 52.41, | |
| "learning_rate": 1.1241134751773048e-05, | |
| "loss": 0.6042, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 53.01, | |
| "learning_rate": 1.0354609929078014e-05, | |
| "loss": 0.5915, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 53.01, | |
| "eval_loss": 0.3120293617248535, | |
| "eval_runtime": 85.3554, | |
| "eval_samples_per_second": 25.974, | |
| "eval_steps_per_second": 1.628, | |
| "eval_wer": 0.26806153846153846, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 53.61, | |
| "learning_rate": 9.468085106382977e-06, | |
| "loss": 0.5774, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 54.22, | |
| "learning_rate": 8.581560283687943e-06, | |
| "loss": 0.5936, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 54.82, | |
| "learning_rate": 7.695035460992907e-06, | |
| "loss": 0.5876, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 55.42, | |
| "learning_rate": 6.808510638297872e-06, | |
| "loss": 0.5798, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 55.42, | |
| "eval_loss": 0.3132975995540619, | |
| "eval_runtime": 86.5587, | |
| "eval_samples_per_second": 25.613, | |
| "eval_steps_per_second": 1.606, | |
| "eval_wer": 0.2708923076923077, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 56.02, | |
| "learning_rate": 5.921985815602837e-06, | |
| "loss": 0.5921, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 56.63, | |
| "learning_rate": 5.035460992907801e-06, | |
| "loss": 0.5603, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 57.23, | |
| "learning_rate": 4.148936170212765e-06, | |
| "loss": 0.5659, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 57.83, | |
| "learning_rate": 3.2624113475177303e-06, | |
| "loss": 0.5736, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 57.83, | |
| "eval_loss": 0.3086037039756775, | |
| "eval_runtime": 85.004, | |
| "eval_samples_per_second": 26.081, | |
| "eval_steps_per_second": 1.635, | |
| "eval_wer": 0.2675692307692308, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 58.43, | |
| "learning_rate": 2.3758865248226945e-06, | |
| "loss": 0.5714, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 59.04, | |
| "learning_rate": 1.4893617021276596e-06, | |
| "loss": 0.5835, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 59.64, | |
| "learning_rate": 6.02836879432624e-07, | |
| "loss": 0.5691, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "step": 9960, | |
| "total_flos": 3.851174581370918e+19, | |
| "train_loss": 1.4761058049029614, | |
| "train_runtime": 17861.6167, | |
| "train_samples_per_second": 17.74, | |
| "train_steps_per_second": 0.558 | |
| } | |
| ], | |
| "max_steps": 9960, | |
| "num_train_epochs": 60, | |
| "total_flos": 3.851174581370918e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |