|
{ |
|
"best_metric": 2.100327730178833, |
|
"best_model_checkpoint": "./drive/MyDrive/repositories/torch_example_image-classification/outputs_letter3/checkpoint-1043", |
|
"epoch": 150.0, |
|
"eval_steps": 500, |
|
"global_step": 1050, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.015384615384615385, |
|
"eval_loss": 3.9449453353881836, |
|
"eval_runtime": 2.0983, |
|
"eval_samples_per_second": 61.955, |
|
"eval_steps_per_second": 0.953, |
|
"step": 7 |
|
}, |
|
{ |
|
"epoch": 1.43, |
|
"learning_rate": 1.9714285714285718e-05, |
|
"loss": 3.9333, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.023076923076923078, |
|
"eval_loss": 3.9366512298583984, |
|
"eval_runtime": 2.2606, |
|
"eval_samples_per_second": 57.506, |
|
"eval_steps_per_second": 0.885, |
|
"step": 14 |
|
}, |
|
{ |
|
"epoch": 2.86, |
|
"learning_rate": 1.942857142857143e-05, |
|
"loss": 3.8939, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.03076923076923077, |
|
"eval_loss": 3.9279532432556152, |
|
"eval_runtime": 2.3275, |
|
"eval_samples_per_second": 55.855, |
|
"eval_steps_per_second": 0.859, |
|
"step": 21 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.046153846153846156, |
|
"eval_loss": 3.9167158603668213, |
|
"eval_runtime": 2.2361, |
|
"eval_samples_per_second": 58.137, |
|
"eval_steps_per_second": 0.894, |
|
"step": 28 |
|
}, |
|
{ |
|
"epoch": 4.29, |
|
"learning_rate": 1.9142857142857146e-05, |
|
"loss": 3.8562, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_accuracy": 0.06923076923076923, |
|
"eval_loss": 3.9032633304595947, |
|
"eval_runtime": 2.4923, |
|
"eval_samples_per_second": 52.161, |
|
"eval_steps_per_second": 0.802, |
|
"step": 35 |
|
}, |
|
{ |
|
"epoch": 5.71, |
|
"learning_rate": 1.885714285714286e-05, |
|
"loss": 3.8008, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_accuracy": 0.07692307692307693, |
|
"eval_loss": 3.8873634338378906, |
|
"eval_runtime": 2.5262, |
|
"eval_samples_per_second": 51.461, |
|
"eval_steps_per_second": 0.792, |
|
"step": 42 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_accuracy": 0.1076923076923077, |
|
"eval_loss": 3.867002010345459, |
|
"eval_runtime": 2.5898, |
|
"eval_samples_per_second": 50.196, |
|
"eval_steps_per_second": 0.772, |
|
"step": 49 |
|
}, |
|
{ |
|
"epoch": 7.14, |
|
"learning_rate": 1.8571428571428575e-05, |
|
"loss": 3.7555, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_accuracy": 0.1, |
|
"eval_loss": 3.8494887351989746, |
|
"eval_runtime": 2.6867, |
|
"eval_samples_per_second": 48.387, |
|
"eval_steps_per_second": 0.744, |
|
"step": 56 |
|
}, |
|
{ |
|
"epoch": 8.57, |
|
"learning_rate": 1.8285714285714288e-05, |
|
"loss": 3.6917, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_accuracy": 0.11538461538461539, |
|
"eval_loss": 3.8304965496063232, |
|
"eval_runtime": 2.4963, |
|
"eval_samples_per_second": 52.077, |
|
"eval_steps_per_second": 0.801, |
|
"step": 63 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"learning_rate": 1.8e-05, |
|
"loss": 3.6372, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"eval_accuracy": 0.13846153846153847, |
|
"eval_loss": 3.8138267993927, |
|
"eval_runtime": 2.3942, |
|
"eval_samples_per_second": 54.298, |
|
"eval_steps_per_second": 0.835, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 11.0, |
|
"eval_accuracy": 0.12307692307692308, |
|
"eval_loss": 3.796628952026367, |
|
"eval_runtime": 2.3929, |
|
"eval_samples_per_second": 54.328, |
|
"eval_steps_per_second": 0.836, |
|
"step": 77 |
|
}, |
|
{ |
|
"epoch": 11.43, |
|
"learning_rate": 1.7714285714285717e-05, |
|
"loss": 3.5846, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 12.0, |
|
"eval_accuracy": 0.15384615384615385, |
|
"eval_loss": 3.7766666412353516, |
|
"eval_runtime": 2.4605, |
|
"eval_samples_per_second": 52.835, |
|
"eval_steps_per_second": 0.813, |
|
"step": 84 |
|
}, |
|
{ |
|
"epoch": 12.86, |
|
"learning_rate": 1.742857142857143e-05, |
|
"loss": 3.5047, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 13.0, |
|
"eval_accuracy": 0.23076923076923078, |
|
"eval_loss": 3.751601219177246, |
|
"eval_runtime": 2.5029, |
|
"eval_samples_per_second": 51.94, |
|
"eval_steps_per_second": 0.799, |
|
"step": 91 |
|
}, |
|
{ |
|
"epoch": 14.0, |
|
"eval_accuracy": 0.23846153846153847, |
|
"eval_loss": 3.7278525829315186, |
|
"eval_runtime": 2.6158, |
|
"eval_samples_per_second": 49.699, |
|
"eval_steps_per_second": 0.765, |
|
"step": 98 |
|
}, |
|
{ |
|
"epoch": 14.29, |
|
"learning_rate": 1.7142857142857142e-05, |
|
"loss": 3.4547, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 15.0, |
|
"eval_accuracy": 0.23846153846153847, |
|
"eval_loss": 3.7031078338623047, |
|
"eval_runtime": 2.5457, |
|
"eval_samples_per_second": 51.066, |
|
"eval_steps_per_second": 0.786, |
|
"step": 105 |
|
}, |
|
{ |
|
"epoch": 15.71, |
|
"learning_rate": 1.6857142857142858e-05, |
|
"loss": 3.3796, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 16.0, |
|
"eval_accuracy": 0.2692307692307692, |
|
"eval_loss": 3.672494649887085, |
|
"eval_runtime": 2.368, |
|
"eval_samples_per_second": 54.899, |
|
"eval_steps_per_second": 0.845, |
|
"step": 112 |
|
}, |
|
{ |
|
"epoch": 17.0, |
|
"eval_accuracy": 0.27692307692307694, |
|
"eval_loss": 3.646197557449341, |
|
"eval_runtime": 2.3752, |
|
"eval_samples_per_second": 54.731, |
|
"eval_steps_per_second": 0.842, |
|
"step": 119 |
|
}, |
|
{ |
|
"epoch": 17.14, |
|
"learning_rate": 1.6571428571428574e-05, |
|
"loss": 3.3283, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 18.0, |
|
"eval_accuracy": 0.2923076923076923, |
|
"eval_loss": 3.622591495513916, |
|
"eval_runtime": 2.3748, |
|
"eval_samples_per_second": 54.741, |
|
"eval_steps_per_second": 0.842, |
|
"step": 126 |
|
}, |
|
{ |
|
"epoch": 18.57, |
|
"learning_rate": 1.6285714285714287e-05, |
|
"loss": 3.2728, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 19.0, |
|
"eval_accuracy": 0.2846153846153846, |
|
"eval_loss": 3.6022486686706543, |
|
"eval_runtime": 2.4033, |
|
"eval_samples_per_second": 54.091, |
|
"eval_steps_per_second": 0.832, |
|
"step": 133 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"learning_rate": 1.6000000000000003e-05, |
|
"loss": 3.2229, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"eval_accuracy": 0.27692307692307694, |
|
"eval_loss": 3.5929768085479736, |
|
"eval_runtime": 2.4687, |
|
"eval_samples_per_second": 52.66, |
|
"eval_steps_per_second": 0.81, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 21.0, |
|
"eval_accuracy": 0.33076923076923076, |
|
"eval_loss": 3.574838876724243, |
|
"eval_runtime": 2.5603, |
|
"eval_samples_per_second": 50.775, |
|
"eval_steps_per_second": 0.781, |
|
"step": 147 |
|
}, |
|
{ |
|
"epoch": 21.43, |
|
"learning_rate": 1.5714285714285715e-05, |
|
"loss": 3.1514, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 22.0, |
|
"eval_accuracy": 0.3384615384615385, |
|
"eval_loss": 3.5403976440429688, |
|
"eval_runtime": 2.3284, |
|
"eval_samples_per_second": 55.833, |
|
"eval_steps_per_second": 0.859, |
|
"step": 154 |
|
}, |
|
{ |
|
"epoch": 22.86, |
|
"learning_rate": 1.542857142857143e-05, |
|
"loss": 3.1179, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 23.0, |
|
"eval_accuracy": 0.3384615384615385, |
|
"eval_loss": 3.514573335647583, |
|
"eval_runtime": 2.3901, |
|
"eval_samples_per_second": 54.39, |
|
"eval_steps_per_second": 0.837, |
|
"step": 161 |
|
}, |
|
{ |
|
"epoch": 24.0, |
|
"eval_accuracy": 0.34615384615384615, |
|
"eval_loss": 3.491649866104126, |
|
"eval_runtime": 2.3495, |
|
"eval_samples_per_second": 55.331, |
|
"eval_steps_per_second": 0.851, |
|
"step": 168 |
|
}, |
|
{ |
|
"epoch": 24.29, |
|
"learning_rate": 1.5142857142857144e-05, |
|
"loss": 3.0559, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 25.0, |
|
"eval_accuracy": 0.3384615384615385, |
|
"eval_loss": 3.473316192626953, |
|
"eval_runtime": 2.387, |
|
"eval_samples_per_second": 54.461, |
|
"eval_steps_per_second": 0.838, |
|
"step": 175 |
|
}, |
|
{ |
|
"epoch": 25.71, |
|
"learning_rate": 1.4857142857142858e-05, |
|
"loss": 3.0051, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 26.0, |
|
"eval_accuracy": 0.36153846153846153, |
|
"eval_loss": 3.453972339630127, |
|
"eval_runtime": 2.5367, |
|
"eval_samples_per_second": 51.249, |
|
"eval_steps_per_second": 0.788, |
|
"step": 182 |
|
}, |
|
{ |
|
"epoch": 27.0, |
|
"eval_accuracy": 0.36923076923076925, |
|
"eval_loss": 3.4498963356018066, |
|
"eval_runtime": 2.5601, |
|
"eval_samples_per_second": 50.779, |
|
"eval_steps_per_second": 0.781, |
|
"step": 189 |
|
}, |
|
{ |
|
"epoch": 27.14, |
|
"learning_rate": 1.4571428571428573e-05, |
|
"loss": 2.9775, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 28.0, |
|
"eval_accuracy": 0.3769230769230769, |
|
"eval_loss": 3.435525417327881, |
|
"eval_runtime": 2.3663, |
|
"eval_samples_per_second": 54.939, |
|
"eval_steps_per_second": 0.845, |
|
"step": 196 |
|
}, |
|
{ |
|
"epoch": 28.57, |
|
"learning_rate": 1.4285714285714287e-05, |
|
"loss": 2.9277, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 29.0, |
|
"eval_accuracy": 0.38461538461538464, |
|
"eval_loss": 3.4166250228881836, |
|
"eval_runtime": 2.3557, |
|
"eval_samples_per_second": 55.185, |
|
"eval_steps_per_second": 0.849, |
|
"step": 203 |
|
}, |
|
{ |
|
"epoch": 30.0, |
|
"learning_rate": 1.4e-05, |
|
"loss": 2.9066, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 30.0, |
|
"eval_accuracy": 0.4, |
|
"eval_loss": 3.40069842338562, |
|
"eval_runtime": 2.3914, |
|
"eval_samples_per_second": 54.362, |
|
"eval_steps_per_second": 0.836, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 31.0, |
|
"eval_accuracy": 0.36923076923076925, |
|
"eval_loss": 3.3825573921203613, |
|
"eval_runtime": 2.5239, |
|
"eval_samples_per_second": 51.509, |
|
"eval_steps_per_second": 0.792, |
|
"step": 217 |
|
}, |
|
{ |
|
"epoch": 31.43, |
|
"learning_rate": 1.3714285714285716e-05, |
|
"loss": 2.8464, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 32.0, |
|
"eval_accuracy": 0.4076923076923077, |
|
"eval_loss": 3.369760274887085, |
|
"eval_runtime": 2.3953, |
|
"eval_samples_per_second": 54.273, |
|
"eval_steps_per_second": 0.835, |
|
"step": 224 |
|
}, |
|
{ |
|
"epoch": 32.86, |
|
"learning_rate": 1.3428571428571429e-05, |
|
"loss": 2.8044, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 33.0, |
|
"eval_accuracy": 0.4076923076923077, |
|
"eval_loss": 3.350893497467041, |
|
"eval_runtime": 2.5141, |
|
"eval_samples_per_second": 51.708, |
|
"eval_steps_per_second": 0.796, |
|
"step": 231 |
|
}, |
|
{ |
|
"epoch": 34.0, |
|
"eval_accuracy": 0.3769230769230769, |
|
"eval_loss": 3.324298620223999, |
|
"eval_runtime": 2.5107, |
|
"eval_samples_per_second": 51.778, |
|
"eval_steps_per_second": 0.797, |
|
"step": 238 |
|
}, |
|
{ |
|
"epoch": 34.29, |
|
"learning_rate": 1.3142857142857145e-05, |
|
"loss": 2.7699, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 35.0, |
|
"eval_accuracy": 0.3923076923076923, |
|
"eval_loss": 3.320146322250366, |
|
"eval_runtime": 2.3695, |
|
"eval_samples_per_second": 54.864, |
|
"eval_steps_per_second": 0.844, |
|
"step": 245 |
|
}, |
|
{ |
|
"epoch": 35.71, |
|
"learning_rate": 1.2857142857142859e-05, |
|
"loss": 2.7251, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 36.0, |
|
"eval_accuracy": 0.4, |
|
"eval_loss": 3.301337242126465, |
|
"eval_runtime": 2.3948, |
|
"eval_samples_per_second": 54.285, |
|
"eval_steps_per_second": 0.835, |
|
"step": 252 |
|
}, |
|
{ |
|
"epoch": 37.0, |
|
"eval_accuracy": 0.4230769230769231, |
|
"eval_loss": 3.2935874462127686, |
|
"eval_runtime": 2.3921, |
|
"eval_samples_per_second": 54.345, |
|
"eval_steps_per_second": 0.836, |
|
"step": 259 |
|
}, |
|
{ |
|
"epoch": 37.14, |
|
"learning_rate": 1.2571428571428572e-05, |
|
"loss": 2.6915, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 38.0, |
|
"eval_accuracy": 0.45384615384615384, |
|
"eval_loss": 3.2827184200286865, |
|
"eval_runtime": 2.4835, |
|
"eval_samples_per_second": 52.345, |
|
"eval_steps_per_second": 0.805, |
|
"step": 266 |
|
}, |
|
{ |
|
"epoch": 38.57, |
|
"learning_rate": 1.2285714285714288e-05, |
|
"loss": 2.6527, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 39.0, |
|
"eval_accuracy": 0.46153846153846156, |
|
"eval_loss": 3.262709379196167, |
|
"eval_runtime": 2.5052, |
|
"eval_samples_per_second": 51.892, |
|
"eval_steps_per_second": 0.798, |
|
"step": 273 |
|
}, |
|
{ |
|
"epoch": 40.0, |
|
"learning_rate": 1.2e-05, |
|
"loss": 2.6541, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 40.0, |
|
"eval_accuracy": 0.46153846153846156, |
|
"eval_loss": 3.2581355571746826, |
|
"eval_runtime": 2.5207, |
|
"eval_samples_per_second": 51.573, |
|
"eval_steps_per_second": 0.793, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 41.0, |
|
"eval_accuracy": 0.4230769230769231, |
|
"eval_loss": 3.2342007160186768, |
|
"eval_runtime": 2.363, |
|
"eval_samples_per_second": 55.016, |
|
"eval_steps_per_second": 0.846, |
|
"step": 287 |
|
}, |
|
{ |
|
"epoch": 41.43, |
|
"learning_rate": 1.1714285714285716e-05, |
|
"loss": 2.5968, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 42.0, |
|
"eval_accuracy": 0.43846153846153846, |
|
"eval_loss": 3.2211241722106934, |
|
"eval_runtime": 2.5033, |
|
"eval_samples_per_second": 51.932, |
|
"eval_steps_per_second": 0.799, |
|
"step": 294 |
|
}, |
|
{ |
|
"epoch": 42.86, |
|
"learning_rate": 1.1428571428571429e-05, |
|
"loss": 2.573, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 43.0, |
|
"eval_accuracy": 0.4076923076923077, |
|
"eval_loss": 3.212242603302002, |
|
"eval_runtime": 2.4013, |
|
"eval_samples_per_second": 54.137, |
|
"eval_steps_per_second": 0.833, |
|
"step": 301 |
|
}, |
|
{ |
|
"epoch": 44.0, |
|
"eval_accuracy": 0.46153846153846156, |
|
"eval_loss": 3.2259228229522705, |
|
"eval_runtime": 2.5062, |
|
"eval_samples_per_second": 51.871, |
|
"eval_steps_per_second": 0.798, |
|
"step": 308 |
|
}, |
|
{ |
|
"epoch": 44.29, |
|
"learning_rate": 1.1142857142857143e-05, |
|
"loss": 2.554, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 45.0, |
|
"eval_accuracy": 0.4307692307692308, |
|
"eval_loss": 3.2271060943603516, |
|
"eval_runtime": 2.357, |
|
"eval_samples_per_second": 55.154, |
|
"eval_steps_per_second": 0.849, |
|
"step": 315 |
|
}, |
|
{ |
|
"epoch": 45.71, |
|
"learning_rate": 1.0857142857142858e-05, |
|
"loss": 2.5222, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 46.0, |
|
"eval_accuracy": 0.4461538461538462, |
|
"eval_loss": 3.2208163738250732, |
|
"eval_runtime": 2.7528, |
|
"eval_samples_per_second": 47.225, |
|
"eval_steps_per_second": 0.727, |
|
"step": 322 |
|
}, |
|
{ |
|
"epoch": 47.0, |
|
"eval_accuracy": 0.4461538461538462, |
|
"eval_loss": 3.2139155864715576, |
|
"eval_runtime": 2.3673, |
|
"eval_samples_per_second": 54.915, |
|
"eval_steps_per_second": 0.845, |
|
"step": 329 |
|
}, |
|
{ |
|
"epoch": 47.14, |
|
"learning_rate": 1.0571428571428572e-05, |
|
"loss": 2.5085, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 48.0, |
|
"eval_accuracy": 0.45384615384615384, |
|
"eval_loss": 3.204026460647583, |
|
"eval_runtime": 2.3343, |
|
"eval_samples_per_second": 55.691, |
|
"eval_steps_per_second": 0.857, |
|
"step": 336 |
|
}, |
|
{ |
|
"epoch": 48.57, |
|
"learning_rate": 1.0285714285714285e-05, |
|
"loss": 2.4593, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 49.0, |
|
"eval_accuracy": 0.49230769230769234, |
|
"eval_loss": 3.205333948135376, |
|
"eval_runtime": 2.3891, |
|
"eval_samples_per_second": 54.414, |
|
"eval_steps_per_second": 0.837, |
|
"step": 343 |
|
}, |
|
{ |
|
"epoch": 50.0, |
|
"learning_rate": 1e-05, |
|
"loss": 2.4585, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 50.0, |
|
"eval_accuracy": 0.47692307692307695, |
|
"eval_loss": 3.1822495460510254, |
|
"eval_runtime": 2.3305, |
|
"eval_samples_per_second": 55.781, |
|
"eval_steps_per_second": 0.858, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 51.0, |
|
"eval_accuracy": 0.46923076923076923, |
|
"eval_loss": 3.1696622371673584, |
|
"eval_runtime": 2.3113, |
|
"eval_samples_per_second": 56.246, |
|
"eval_steps_per_second": 0.865, |
|
"step": 357 |
|
}, |
|
{ |
|
"epoch": 51.43, |
|
"learning_rate": 9.714285714285715e-06, |
|
"loss": 2.4228, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 52.0, |
|
"eval_accuracy": 0.46923076923076923, |
|
"eval_loss": 3.1589303016662598, |
|
"eval_runtime": 2.5283, |
|
"eval_samples_per_second": 51.419, |
|
"eval_steps_per_second": 0.791, |
|
"step": 364 |
|
}, |
|
{ |
|
"epoch": 52.86, |
|
"learning_rate": 9.42857142857143e-06, |
|
"loss": 2.3954, |
|
"step": 370 |
|
}, |
|
{ |
|
"epoch": 53.0, |
|
"eval_accuracy": 0.47692307692307695, |
|
"eval_loss": 3.1375324726104736, |
|
"eval_runtime": 2.3577, |
|
"eval_samples_per_second": 55.139, |
|
"eval_steps_per_second": 0.848, |
|
"step": 371 |
|
}, |
|
{ |
|
"epoch": 54.0, |
|
"eval_accuracy": 0.45384615384615384, |
|
"eval_loss": 3.1092171669006348, |
|
"eval_runtime": 2.3799, |
|
"eval_samples_per_second": 54.625, |
|
"eval_steps_per_second": 0.84, |
|
"step": 378 |
|
}, |
|
{ |
|
"epoch": 54.29, |
|
"learning_rate": 9.142857142857144e-06, |
|
"loss": 2.3641, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 55.0, |
|
"eval_accuracy": 0.47692307692307695, |
|
"eval_loss": 3.099919319152832, |
|
"eval_runtime": 2.3488, |
|
"eval_samples_per_second": 55.348, |
|
"eval_steps_per_second": 0.852, |
|
"step": 385 |
|
}, |
|
{ |
|
"epoch": 55.71, |
|
"learning_rate": 8.857142857142858e-06, |
|
"loss": 2.3651, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 56.0, |
|
"eval_accuracy": 0.46153846153846156, |
|
"eval_loss": 3.0859622955322266, |
|
"eval_runtime": 2.5127, |
|
"eval_samples_per_second": 51.737, |
|
"eval_steps_per_second": 0.796, |
|
"step": 392 |
|
}, |
|
{ |
|
"epoch": 57.0, |
|
"eval_accuracy": 0.46153846153846156, |
|
"eval_loss": 3.0813159942626953, |
|
"eval_runtime": 2.449, |
|
"eval_samples_per_second": 53.082, |
|
"eval_steps_per_second": 0.817, |
|
"step": 399 |
|
}, |
|
{ |
|
"epoch": 57.14, |
|
"learning_rate": 8.571428571428571e-06, |
|
"loss": 2.3182, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 58.0, |
|
"eval_accuracy": 0.49230769230769234, |
|
"eval_loss": 3.0692226886749268, |
|
"eval_runtime": 2.3556, |
|
"eval_samples_per_second": 55.188, |
|
"eval_steps_per_second": 0.849, |
|
"step": 406 |
|
}, |
|
{ |
|
"epoch": 58.57, |
|
"learning_rate": 8.285714285714287e-06, |
|
"loss": 2.3029, |
|
"step": 410 |
|
}, |
|
{ |
|
"epoch": 59.0, |
|
"eval_accuracy": 0.4846153846153846, |
|
"eval_loss": 3.061020612716675, |
|
"eval_runtime": 2.4546, |
|
"eval_samples_per_second": 52.961, |
|
"eval_steps_per_second": 0.815, |
|
"step": 413 |
|
}, |
|
{ |
|
"epoch": 60.0, |
|
"learning_rate": 8.000000000000001e-06, |
|
"loss": 2.2988, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 60.0, |
|
"eval_accuracy": 0.46153846153846156, |
|
"eval_loss": 3.0626943111419678, |
|
"eval_runtime": 2.3415, |
|
"eval_samples_per_second": 55.521, |
|
"eval_steps_per_second": 0.854, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 61.0, |
|
"eval_accuracy": 0.46923076923076923, |
|
"eval_loss": 3.051997184753418, |
|
"eval_runtime": 2.492, |
|
"eval_samples_per_second": 52.167, |
|
"eval_steps_per_second": 0.803, |
|
"step": 427 |
|
}, |
|
{ |
|
"epoch": 61.43, |
|
"learning_rate": 7.714285714285716e-06, |
|
"loss": 2.2865, |
|
"step": 430 |
|
}, |
|
{ |
|
"epoch": 62.0, |
|
"eval_accuracy": 0.45384615384615384, |
|
"eval_loss": 3.039459705352783, |
|
"eval_runtime": 2.4542, |
|
"eval_samples_per_second": 52.97, |
|
"eval_steps_per_second": 0.815, |
|
"step": 434 |
|
}, |
|
{ |
|
"epoch": 62.86, |
|
"learning_rate": 7.428571428571429e-06, |
|
"loss": 2.2623, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 63.0, |
|
"eval_accuracy": 0.46153846153846156, |
|
"eval_loss": 3.0356502532958984, |
|
"eval_runtime": 2.3341, |
|
"eval_samples_per_second": 55.697, |
|
"eval_steps_per_second": 0.857, |
|
"step": 441 |
|
}, |
|
{ |
|
"epoch": 64.0, |
|
"eval_accuracy": 0.46153846153846156, |
|
"eval_loss": 3.0332510471343994, |
|
"eval_runtime": 2.3535, |
|
"eval_samples_per_second": 55.236, |
|
"eval_steps_per_second": 0.85, |
|
"step": 448 |
|
}, |
|
{ |
|
"epoch": 64.29, |
|
"learning_rate": 7.1428571428571436e-06, |
|
"loss": 2.2252, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 65.0, |
|
"eval_accuracy": 0.47692307692307695, |
|
"eval_loss": 3.022916316986084, |
|
"eval_runtime": 2.5362, |
|
"eval_samples_per_second": 51.258, |
|
"eval_steps_per_second": 0.789, |
|
"step": 455 |
|
}, |
|
{ |
|
"epoch": 65.71, |
|
"learning_rate": 6.857142857142858e-06, |
|
"loss": 2.2339, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 66.0, |
|
"eval_accuracy": 0.47692307692307695, |
|
"eval_loss": 3.020322799682617, |
|
"eval_runtime": 2.3519, |
|
"eval_samples_per_second": 55.275, |
|
"eval_steps_per_second": 0.85, |
|
"step": 462 |
|
}, |
|
{ |
|
"epoch": 67.0, |
|
"eval_accuracy": 0.49230769230769234, |
|
"eval_loss": 3.0076119899749756, |
|
"eval_runtime": 2.359, |
|
"eval_samples_per_second": 55.107, |
|
"eval_steps_per_second": 0.848, |
|
"step": 469 |
|
}, |
|
{ |
|
"epoch": 67.14, |
|
"learning_rate": 6.571428571428572e-06, |
|
"loss": 2.2017, |
|
"step": 470 |
|
}, |
|
{ |
|
"epoch": 68.0, |
|
"eval_accuracy": 0.4846153846153846, |
|
"eval_loss": 2.9876248836517334, |
|
"eval_runtime": 2.372, |
|
"eval_samples_per_second": 54.805, |
|
"eval_steps_per_second": 0.843, |
|
"step": 476 |
|
}, |
|
{ |
|
"epoch": 68.57, |
|
"learning_rate": 6.285714285714286e-06, |
|
"loss": 2.1972, |
|
"step": 480 |
|
}, |
|
{ |
|
"epoch": 69.0, |
|
"eval_accuracy": 0.49230769230769234, |
|
"eval_loss": 2.971572160720825, |
|
"eval_runtime": 2.3479, |
|
"eval_samples_per_second": 55.368, |
|
"eval_steps_per_second": 0.852, |
|
"step": 483 |
|
}, |
|
{ |
|
"epoch": 70.0, |
|
"learning_rate": 6e-06, |
|
"loss": 2.1964, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 70.0, |
|
"eval_accuracy": 0.5, |
|
"eval_loss": 2.96323823928833, |
|
"eval_runtime": 2.4351, |
|
"eval_samples_per_second": 53.387, |
|
"eval_steps_per_second": 0.821, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 71.0, |
|
"eval_accuracy": 0.49230769230769234, |
|
"eval_loss": 2.9597153663635254, |
|
"eval_runtime": 2.4929, |
|
"eval_samples_per_second": 52.149, |
|
"eval_steps_per_second": 0.802, |
|
"step": 497 |
|
}, |
|
{ |
|
"epoch": 71.43, |
|
"learning_rate": 5.7142857142857145e-06, |
|
"loss": 2.1775, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 72.0, |
|
"eval_accuracy": 0.5, |
|
"eval_loss": 2.9580914974212646, |
|
"eval_runtime": 2.5027, |
|
"eval_samples_per_second": 51.943, |
|
"eval_steps_per_second": 0.799, |
|
"step": 504 |
|
}, |
|
{ |
|
"epoch": 72.86, |
|
"learning_rate": 5.428571428571429e-06, |
|
"loss": 2.1619, |
|
"step": 510 |
|
}, |
|
{ |
|
"epoch": 73.0, |
|
"eval_accuracy": 0.5076923076923077, |
|
"eval_loss": 2.9516122341156006, |
|
"eval_runtime": 2.4828, |
|
"eval_samples_per_second": 52.361, |
|
"eval_steps_per_second": 0.806, |
|
"step": 511 |
|
}, |
|
{ |
|
"epoch": 74.0, |
|
"eval_accuracy": 0.5153846153846153, |
|
"eval_loss": 2.935607433319092, |
|
"eval_runtime": 2.3758, |
|
"eval_samples_per_second": 54.719, |
|
"eval_steps_per_second": 0.842, |
|
"step": 518 |
|
}, |
|
{ |
|
"epoch": 74.29, |
|
"learning_rate": 5.142857142857142e-06, |
|
"loss": 2.1633, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 75.0, |
|
"eval_accuracy": 0.5076923076923077, |
|
"eval_loss": 2.928561210632324, |
|
"eval_runtime": 2.3327, |
|
"eval_samples_per_second": 55.729, |
|
"eval_steps_per_second": 0.857, |
|
"step": 525 |
|
}, |
|
{ |
|
"epoch": 75.71, |
|
"learning_rate": 4.857142857142858e-06, |
|
"loss": 2.1207, |
|
"step": 530 |
|
}, |
|
{ |
|
"epoch": 76.0, |
|
"eval_accuracy": 0.5153846153846153, |
|
"eval_loss": 2.9266109466552734, |
|
"eval_runtime": 2.3813, |
|
"eval_samples_per_second": 54.591, |
|
"eval_steps_per_second": 0.84, |
|
"step": 532 |
|
}, |
|
{ |
|
"epoch": 77.0, |
|
"eval_accuracy": 0.5230769230769231, |
|
"eval_loss": 2.920504093170166, |
|
"eval_runtime": 2.4777, |
|
"eval_samples_per_second": 52.468, |
|
"eval_steps_per_second": 0.807, |
|
"step": 539 |
|
}, |
|
{ |
|
"epoch": 77.14, |
|
"learning_rate": 4.571428571428572e-06, |
|
"loss": 2.1353, |
|
"step": 540 |
|
}, |
|
{ |
|
"epoch": 78.0, |
|
"eval_accuracy": 0.5153846153846153, |
|
"eval_loss": 2.9131386280059814, |
|
"eval_runtime": 2.3865, |
|
"eval_samples_per_second": 54.473, |
|
"eval_steps_per_second": 0.838, |
|
"step": 546 |
|
}, |
|
{ |
|
"epoch": 78.57, |
|
"learning_rate": 4.2857142857142855e-06, |
|
"loss": 2.1075, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 79.0, |
|
"eval_accuracy": 0.5230769230769231, |
|
"eval_loss": 2.907454252243042, |
|
"eval_runtime": 2.3713, |
|
"eval_samples_per_second": 54.823, |
|
"eval_steps_per_second": 0.843, |
|
"step": 553 |
|
}, |
|
{ |
|
"epoch": 80.0, |
|
"learning_rate": 4.000000000000001e-06, |
|
"loss": 2.1025, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 80.0, |
|
"eval_accuracy": 0.5230769230769231, |
|
"eval_loss": 2.907280921936035, |
|
"eval_runtime": 2.3736, |
|
"eval_samples_per_second": 54.768, |
|
"eval_steps_per_second": 0.843, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 81.0, |
|
"eval_accuracy": 0.5153846153846153, |
|
"eval_loss": 2.9174439907073975, |
|
"eval_runtime": 2.3602, |
|
"eval_samples_per_second": 55.079, |
|
"eval_steps_per_second": 0.847, |
|
"step": 567 |
|
}, |
|
{ |
|
"epoch": 81.43, |
|
"learning_rate": 3.7142857142857146e-06, |
|
"loss": 2.1031, |
|
"step": 570 |
|
}, |
|
{ |
|
"epoch": 82.0, |
|
"eval_accuracy": 0.5307692307692308, |
|
"eval_loss": 2.913137197494507, |
|
"eval_runtime": 2.4712, |
|
"eval_samples_per_second": 52.605, |
|
"eval_steps_per_second": 0.809, |
|
"step": 574 |
|
}, |
|
{ |
|
"epoch": 82.86, |
|
"learning_rate": 3.428571428571429e-06, |
|
"loss": 2.0932, |
|
"step": 580 |
|
}, |
|
{ |
|
"epoch": 83.0, |
|
"eval_accuracy": 0.5307692307692308, |
|
"eval_loss": 2.909245491027832, |
|
"eval_runtime": 2.3787, |
|
"eval_samples_per_second": 54.651, |
|
"eval_steps_per_second": 0.841, |
|
"step": 581 |
|
}, |
|
{ |
|
"epoch": 84.0, |
|
"eval_accuracy": 0.5307692307692308, |
|
"eval_loss": 2.8977668285369873, |
|
"eval_runtime": 4.0301, |
|
"eval_samples_per_second": 32.257, |
|
"eval_steps_per_second": 0.496, |
|
"step": 588 |
|
}, |
|
{ |
|
"epoch": 84.29, |
|
"learning_rate": 3.142857142857143e-06, |
|
"loss": 2.0861, |
|
"step": 590 |
|
}, |
|
{ |
|
"epoch": 85.0, |
|
"eval_accuracy": 0.5307692307692308, |
|
"eval_loss": 2.8871328830718994, |
|
"eval_runtime": 3.7091, |
|
"eval_samples_per_second": 35.049, |
|
"eval_steps_per_second": 0.539, |
|
"step": 595 |
|
}, |
|
{ |
|
"epoch": 85.71, |
|
"learning_rate": 2.8571428571428573e-06, |
|
"loss": 2.0478, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 86.0, |
|
"eval_accuracy": 0.5384615384615384, |
|
"eval_loss": 2.882897138595581, |
|
"eval_runtime": 4.3893, |
|
"eval_samples_per_second": 29.617, |
|
"eval_steps_per_second": 0.456, |
|
"step": 602 |
|
}, |
|
{ |
|
"epoch": 87.0, |
|
"eval_accuracy": 0.5461538461538461, |
|
"eval_loss": 2.880449056625366, |
|
"eval_runtime": 2.9082, |
|
"eval_samples_per_second": 44.701, |
|
"eval_steps_per_second": 0.688, |
|
"step": 609 |
|
}, |
|
{ |
|
"epoch": 87.14, |
|
"learning_rate": 2.571428571428571e-06, |
|
"loss": 2.0815, |
|
"step": 610 |
|
}, |
|
{ |
|
"epoch": 88.0, |
|
"eval_accuracy": 0.5461538461538461, |
|
"eval_loss": 2.872459888458252, |
|
"eval_runtime": 2.8921, |
|
"eval_samples_per_second": 44.95, |
|
"eval_steps_per_second": 0.692, |
|
"step": 616 |
|
}, |
|
{ |
|
"epoch": 88.57, |
|
"learning_rate": 2.285714285714286e-06, |
|
"loss": 2.0756, |
|
"step": 620 |
|
}, |
|
{ |
|
"epoch": 89.0, |
|
"eval_accuracy": 0.5461538461538461, |
|
"eval_loss": 2.8694419860839844, |
|
"eval_runtime": 3.9643, |
|
"eval_samples_per_second": 32.793, |
|
"eval_steps_per_second": 0.505, |
|
"step": 623 |
|
}, |
|
{ |
|
"epoch": 90.0, |
|
"learning_rate": 2.0000000000000003e-06, |
|
"loss": 2.065, |
|
"step": 630 |
|
}, |
|
{ |
|
"epoch": 90.0, |
|
"eval_accuracy": 0.5461538461538461, |
|
"eval_loss": 2.866473913192749, |
|
"eval_runtime": 3.1035, |
|
"eval_samples_per_second": 41.888, |
|
"eval_steps_per_second": 0.644, |
|
"step": 630 |
|
}, |
|
{ |
|
"epoch": 91.0, |
|
"eval_accuracy": 0.5461538461538461, |
|
"eval_loss": 2.861470937728882, |
|
"eval_runtime": 3.1207, |
|
"eval_samples_per_second": 41.657, |
|
"eval_steps_per_second": 0.641, |
|
"step": 637 |
|
}, |
|
{ |
|
"epoch": 91.43, |
|
"learning_rate": 1.7142857142857145e-06, |
|
"loss": 2.0572, |
|
"step": 640 |
|
}, |
|
{ |
|
"epoch": 92.0, |
|
"eval_accuracy": 0.5461538461538461, |
|
"eval_loss": 2.859877347946167, |
|
"eval_runtime": 2.6425, |
|
"eval_samples_per_second": 49.197, |
|
"eval_steps_per_second": 0.757, |
|
"step": 644 |
|
}, |
|
{ |
|
"epoch": 92.86, |
|
"learning_rate": 1.4285714285714286e-06, |
|
"loss": 2.0358, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 93.0, |
|
"eval_accuracy": 0.5461538461538461, |
|
"eval_loss": 2.8619918823242188, |
|
"eval_runtime": 2.4174, |
|
"eval_samples_per_second": 53.776, |
|
"eval_steps_per_second": 0.827, |
|
"step": 651 |
|
}, |
|
{ |
|
"epoch": 94.0, |
|
"eval_accuracy": 0.5461538461538461, |
|
"eval_loss": 2.862945079803467, |
|
"eval_runtime": 2.4216, |
|
"eval_samples_per_second": 53.684, |
|
"eval_steps_per_second": 0.826, |
|
"step": 658 |
|
}, |
|
{ |
|
"epoch": 94.29, |
|
"learning_rate": 1.142857142857143e-06, |
|
"loss": 2.0663, |
|
"step": 660 |
|
}, |
|
{ |
|
"epoch": 95.0, |
|
"eval_accuracy": 0.5538461538461539, |
|
"eval_loss": 2.862516403198242, |
|
"eval_runtime": 2.3749, |
|
"eval_samples_per_second": 54.739, |
|
"eval_steps_per_second": 0.842, |
|
"step": 665 |
|
}, |
|
{ |
|
"epoch": 95.71, |
|
"learning_rate": 8.571428571428572e-07, |
|
"loss": 2.0353, |
|
"step": 670 |
|
}, |
|
{ |
|
"epoch": 96.0, |
|
"eval_accuracy": 0.5538461538461539, |
|
"eval_loss": 2.8627583980560303, |
|
"eval_runtime": 2.3527, |
|
"eval_samples_per_second": 55.257, |
|
"eval_steps_per_second": 0.85, |
|
"step": 672 |
|
}, |
|
{ |
|
"epoch": 97.0, |
|
"eval_accuracy": 0.5538461538461539, |
|
"eval_loss": 2.8628671169281006, |
|
"eval_runtime": 2.5748, |
|
"eval_samples_per_second": 50.49, |
|
"eval_steps_per_second": 0.777, |
|
"step": 679 |
|
}, |
|
{ |
|
"epoch": 97.14, |
|
"learning_rate": 5.714285714285715e-07, |
|
"loss": 2.0506, |
|
"step": 680 |
|
}, |
|
{ |
|
"epoch": 98.0, |
|
"eval_accuracy": 0.5538461538461539, |
|
"eval_loss": 2.862170457839966, |
|
"eval_runtime": 2.3987, |
|
"eval_samples_per_second": 54.196, |
|
"eval_steps_per_second": 0.834, |
|
"step": 686 |
|
}, |
|
{ |
|
"epoch": 98.57, |
|
"learning_rate": 2.8571428571428575e-07, |
|
"loss": 2.0494, |
|
"step": 690 |
|
}, |
|
{ |
|
"epoch": 99.0, |
|
"eval_accuracy": 0.5538461538461539, |
|
"eval_loss": 2.8621606826782227, |
|
"eval_runtime": 2.676, |
|
"eval_samples_per_second": 48.58, |
|
"eval_steps_per_second": 0.747, |
|
"step": 693 |
|
}, |
|
{ |
|
"epoch": 100.0, |
|
"learning_rate": 0.0, |
|
"loss": 2.0566, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 100.0, |
|
"eval_accuracy": 0.5538461538461539, |
|
"eval_loss": 2.862189292907715, |
|
"eval_runtime": 2.8132, |
|
"eval_samples_per_second": 46.211, |
|
"eval_steps_per_second": 0.711, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 100.0, |
|
"step": 700, |
|
"total_flos": 4.031389254057984e+18, |
|
"train_loss": 2.6316724504743303, |
|
"train_runtime": 4344.3701, |
|
"train_samples_per_second": 11.97, |
|
"train_steps_per_second": 0.161 |
|
}, |
|
{ |
|
"epoch": 101.0, |
|
"eval_accuracy": 0.5153846153846153, |
|
"eval_loss": 2.84305477142334, |
|
"eval_runtime": 2.1617, |
|
"eval_samples_per_second": 60.137, |
|
"eval_steps_per_second": 0.925, |
|
"step": 707 |
|
}, |
|
{ |
|
"epoch": 101.43, |
|
"learning_rate": 1.9761904761904763e-05, |
|
"loss": 2.067, |
|
"step": 710 |
|
}, |
|
{ |
|
"epoch": 102.0, |
|
"eval_accuracy": 0.5, |
|
"eval_loss": 2.8063366413116455, |
|
"eval_runtime": 2.1563, |
|
"eval_samples_per_second": 60.288, |
|
"eval_steps_per_second": 0.928, |
|
"step": 714 |
|
}, |
|
{ |
|
"epoch": 102.86, |
|
"learning_rate": 1.9523809523809524e-05, |
|
"loss": 2.0234, |
|
"step": 720 |
|
}, |
|
{ |
|
"epoch": 103.0, |
|
"eval_accuracy": 0.5230769230769231, |
|
"eval_loss": 2.7845773696899414, |
|
"eval_runtime": 2.1676, |
|
"eval_samples_per_second": 59.974, |
|
"eval_steps_per_second": 0.923, |
|
"step": 721 |
|
}, |
|
{ |
|
"epoch": 104.0, |
|
"eval_accuracy": 0.5307692307692308, |
|
"eval_loss": 2.762021541595459, |
|
"eval_runtime": 2.2767, |
|
"eval_samples_per_second": 57.1, |
|
"eval_steps_per_second": 0.878, |
|
"step": 728 |
|
}, |
|
{ |
|
"epoch": 104.29, |
|
"learning_rate": 1.928571428571429e-05, |
|
"loss": 1.9979, |
|
"step": 730 |
|
}, |
|
{ |
|
"epoch": 105.0, |
|
"eval_accuracy": 0.5538461538461539, |
|
"eval_loss": 2.7421603202819824, |
|
"eval_runtime": 2.2859, |
|
"eval_samples_per_second": 56.871, |
|
"eval_steps_per_second": 0.875, |
|
"step": 735 |
|
}, |
|
{ |
|
"epoch": 105.71, |
|
"learning_rate": 1.904761904761905e-05, |
|
"loss": 1.9515, |
|
"step": 740 |
|
}, |
|
{ |
|
"epoch": 106.0, |
|
"eval_accuracy": 0.5615384615384615, |
|
"eval_loss": 2.722872734069824, |
|
"eval_runtime": 2.3487, |
|
"eval_samples_per_second": 55.35, |
|
"eval_steps_per_second": 0.852, |
|
"step": 742 |
|
}, |
|
{ |
|
"epoch": 107.0, |
|
"eval_accuracy": 0.5384615384615384, |
|
"eval_loss": 2.7145817279815674, |
|
"eval_runtime": 2.3976, |
|
"eval_samples_per_second": 54.221, |
|
"eval_steps_per_second": 0.834, |
|
"step": 749 |
|
}, |
|
{ |
|
"epoch": 107.14, |
|
"learning_rate": 1.880952380952381e-05, |
|
"loss": 1.9095, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 108.0, |
|
"eval_accuracy": 0.5538461538461539, |
|
"eval_loss": 2.6905264854431152, |
|
"eval_runtime": 2.3887, |
|
"eval_samples_per_second": 54.422, |
|
"eval_steps_per_second": 0.837, |
|
"step": 756 |
|
}, |
|
{ |
|
"epoch": 108.57, |
|
"learning_rate": 1.8571428571428575e-05, |
|
"loss": 1.8759, |
|
"step": 760 |
|
}, |
|
{ |
|
"epoch": 109.0, |
|
"eval_accuracy": 0.5538461538461539, |
|
"eval_loss": 2.6840174198150635, |
|
"eval_runtime": 2.369, |
|
"eval_samples_per_second": 54.876, |
|
"eval_steps_per_second": 0.844, |
|
"step": 763 |
|
}, |
|
{ |
|
"epoch": 110.0, |
|
"learning_rate": 1.8333333333333333e-05, |
|
"loss": 1.8561, |
|
"step": 770 |
|
}, |
|
{ |
|
"epoch": 110.0, |
|
"eval_accuracy": 0.5461538461538461, |
|
"eval_loss": 2.6554973125457764, |
|
"eval_runtime": 2.3964, |
|
"eval_samples_per_second": 54.248, |
|
"eval_steps_per_second": 0.835, |
|
"step": 770 |
|
}, |
|
{ |
|
"epoch": 111.0, |
|
"eval_accuracy": 0.5692307692307692, |
|
"eval_loss": 2.6465163230895996, |
|
"eval_runtime": 2.5418, |
|
"eval_samples_per_second": 51.146, |
|
"eval_steps_per_second": 0.787, |
|
"step": 777 |
|
}, |
|
{ |
|
"epoch": 111.43, |
|
"learning_rate": 1.8095238095238097e-05, |
|
"loss": 1.8147, |
|
"step": 780 |
|
}, |
|
{ |
|
"epoch": 112.0, |
|
"eval_accuracy": 0.5384615384615384, |
|
"eval_loss": 2.6321260929107666, |
|
"eval_runtime": 2.6607, |
|
"eval_samples_per_second": 48.859, |
|
"eval_steps_per_second": 0.752, |
|
"step": 784 |
|
}, |
|
{ |
|
"epoch": 112.86, |
|
"learning_rate": 1.785714285714286e-05, |
|
"loss": 1.799, |
|
"step": 790 |
|
}, |
|
{ |
|
"epoch": 113.0, |
|
"eval_accuracy": 0.5384615384615384, |
|
"eval_loss": 2.6352975368499756, |
|
"eval_runtime": 2.3449, |
|
"eval_samples_per_second": 55.439, |
|
"eval_steps_per_second": 0.853, |
|
"step": 791 |
|
}, |
|
{ |
|
"epoch": 114.0, |
|
"eval_accuracy": 0.5461538461538461, |
|
"eval_loss": 2.6188342571258545, |
|
"eval_runtime": 2.5585, |
|
"eval_samples_per_second": 50.811, |
|
"eval_steps_per_second": 0.782, |
|
"step": 798 |
|
}, |
|
{ |
|
"epoch": 114.29, |
|
"learning_rate": 1.761904761904762e-05, |
|
"loss": 1.7536, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 115.0, |
|
"eval_accuracy": 0.5461538461538461, |
|
"eval_loss": 2.6063857078552246, |
|
"eval_runtime": 2.5441, |
|
"eval_samples_per_second": 51.098, |
|
"eval_steps_per_second": 0.786, |
|
"step": 805 |
|
}, |
|
{ |
|
"epoch": 115.71, |
|
"learning_rate": 1.7380952380952384e-05, |
|
"loss": 1.7162, |
|
"step": 810 |
|
}, |
|
{ |
|
"epoch": 116.0, |
|
"eval_accuracy": 0.5538461538461539, |
|
"eval_loss": 2.591280937194824, |
|
"eval_runtime": 2.5836, |
|
"eval_samples_per_second": 50.316, |
|
"eval_steps_per_second": 0.774, |
|
"step": 812 |
|
}, |
|
{ |
|
"epoch": 117.0, |
|
"eval_accuracy": 0.5923076923076923, |
|
"eval_loss": 2.5749051570892334, |
|
"eval_runtime": 2.5441, |
|
"eval_samples_per_second": 51.099, |
|
"eval_steps_per_second": 0.786, |
|
"step": 819 |
|
}, |
|
{ |
|
"epoch": 117.14, |
|
"learning_rate": 1.7142857142857142e-05, |
|
"loss": 1.7225, |
|
"step": 820 |
|
}, |
|
{ |
|
"epoch": 118.0, |
|
"eval_accuracy": 0.5846153846153846, |
|
"eval_loss": 2.563140869140625, |
|
"eval_runtime": 2.6117, |
|
"eval_samples_per_second": 49.777, |
|
"eval_steps_per_second": 0.766, |
|
"step": 826 |
|
}, |
|
{ |
|
"epoch": 118.57, |
|
"learning_rate": 1.6904761904761906e-05, |
|
"loss": 1.6841, |
|
"step": 830 |
|
}, |
|
{ |
|
"epoch": 119.0, |
|
"eval_accuracy": 0.5923076923076923, |
|
"eval_loss": 2.547883987426758, |
|
"eval_runtime": 2.5689, |
|
"eval_samples_per_second": 50.605, |
|
"eval_steps_per_second": 0.779, |
|
"step": 833 |
|
}, |
|
{ |
|
"epoch": 120.0, |
|
"learning_rate": 1.6666666666666667e-05, |
|
"loss": 1.6388, |
|
"step": 840 |
|
}, |
|
{ |
|
"epoch": 120.0, |
|
"eval_accuracy": 0.5846153846153846, |
|
"eval_loss": 2.5402629375457764, |
|
"eval_runtime": 2.4843, |
|
"eval_samples_per_second": 52.328, |
|
"eval_steps_per_second": 0.805, |
|
"step": 840 |
|
}, |
|
{ |
|
"epoch": 120.0, |
|
"step": 840, |
|
"total_flos": 4.837667104869581e+18, |
|
"train_loss": 0.3072653747740246, |
|
"train_runtime": 892.0889, |
|
"train_samples_per_second": 69.948, |
|
"train_steps_per_second": 0.942 |
|
}, |
|
{ |
|
"epoch": 121.0, |
|
"eval_accuracy": 0.6307692307692307, |
|
"eval_loss": 2.5144028663635254, |
|
"eval_runtime": 2.3105, |
|
"eval_samples_per_second": 56.264, |
|
"eval_steps_per_second": 0.866, |
|
"step": 847 |
|
}, |
|
{ |
|
"epoch": 121.43, |
|
"learning_rate": 1.980952380952381e-05, |
|
"loss": 1.654, |
|
"step": 850 |
|
}, |
|
{ |
|
"epoch": 122.0, |
|
"eval_accuracy": 0.6153846153846154, |
|
"eval_loss": 2.498945474624634, |
|
"eval_runtime": 2.3196, |
|
"eval_samples_per_second": 56.043, |
|
"eval_steps_per_second": 0.862, |
|
"step": 854 |
|
}, |
|
{ |
|
"epoch": 122.86, |
|
"learning_rate": 1.961904761904762e-05, |
|
"loss": 1.5863, |
|
"step": 860 |
|
}, |
|
{ |
|
"epoch": 123.0, |
|
"eval_accuracy": 0.6230769230769231, |
|
"eval_loss": 2.476050853729248, |
|
"eval_runtime": 2.4097, |
|
"eval_samples_per_second": 53.949, |
|
"eval_steps_per_second": 0.83, |
|
"step": 861 |
|
}, |
|
{ |
|
"epoch": 124.0, |
|
"eval_accuracy": 0.6, |
|
"eval_loss": 2.4536540508270264, |
|
"eval_runtime": 2.5779, |
|
"eval_samples_per_second": 50.428, |
|
"eval_steps_per_second": 0.776, |
|
"step": 868 |
|
}, |
|
{ |
|
"epoch": 124.29, |
|
"learning_rate": 1.942857142857143e-05, |
|
"loss": 1.5915, |
|
"step": 870 |
|
}, |
|
{ |
|
"epoch": 125.0, |
|
"eval_accuracy": 0.6076923076923076, |
|
"eval_loss": 2.4367971420288086, |
|
"eval_runtime": 2.4941, |
|
"eval_samples_per_second": 52.123, |
|
"eval_steps_per_second": 0.802, |
|
"step": 875 |
|
}, |
|
{ |
|
"epoch": 125.71, |
|
"learning_rate": 1.923809523809524e-05, |
|
"loss": 1.5221, |
|
"step": 880 |
|
}, |
|
{ |
|
"epoch": 126.0, |
|
"eval_accuracy": 0.5923076923076923, |
|
"eval_loss": 2.4394454956054688, |
|
"eval_runtime": 2.3616, |
|
"eval_samples_per_second": 55.047, |
|
"eval_steps_per_second": 0.847, |
|
"step": 882 |
|
}, |
|
{ |
|
"epoch": 127.0, |
|
"eval_accuracy": 0.6076923076923076, |
|
"eval_loss": 2.3990907669067383, |
|
"eval_runtime": 2.6083, |
|
"eval_samples_per_second": 49.84, |
|
"eval_steps_per_second": 0.767, |
|
"step": 889 |
|
}, |
|
{ |
|
"epoch": 127.14, |
|
"learning_rate": 1.904761904761905e-05, |
|
"loss": 1.5231, |
|
"step": 890 |
|
}, |
|
{ |
|
"epoch": 128.0, |
|
"eval_accuracy": 0.6153846153846154, |
|
"eval_loss": 2.3942794799804688, |
|
"eval_runtime": 2.6252, |
|
"eval_samples_per_second": 49.519, |
|
"eval_steps_per_second": 0.762, |
|
"step": 896 |
|
}, |
|
{ |
|
"epoch": 128.57, |
|
"learning_rate": 1.885714285714286e-05, |
|
"loss": 1.4753, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 129.0, |
|
"eval_accuracy": 0.6307692307692307, |
|
"eval_loss": 2.3880045413970947, |
|
"eval_runtime": 2.4163, |
|
"eval_samples_per_second": 53.801, |
|
"eval_steps_per_second": 0.828, |
|
"step": 903 |
|
}, |
|
{ |
|
"epoch": 130.0, |
|
"learning_rate": 1.866666666666667e-05, |
|
"loss": 1.456, |
|
"step": 910 |
|
}, |
|
{ |
|
"epoch": 130.0, |
|
"eval_accuracy": 0.6230769230769231, |
|
"eval_loss": 2.367373466491699, |
|
"eval_runtime": 2.4274, |
|
"eval_samples_per_second": 53.556, |
|
"eval_steps_per_second": 0.824, |
|
"step": 910 |
|
}, |
|
{ |
|
"epoch": 131.0, |
|
"eval_accuracy": 0.6, |
|
"eval_loss": 2.3696095943450928, |
|
"eval_runtime": 2.3933, |
|
"eval_samples_per_second": 54.319, |
|
"eval_steps_per_second": 0.836, |
|
"step": 917 |
|
}, |
|
{ |
|
"epoch": 131.43, |
|
"learning_rate": 1.8476190476190478e-05, |
|
"loss": 1.4353, |
|
"step": 920 |
|
}, |
|
{ |
|
"epoch": 132.0, |
|
"eval_accuracy": 0.6153846153846154, |
|
"eval_loss": 2.3404488563537598, |
|
"eval_runtime": 2.4129, |
|
"eval_samples_per_second": 53.877, |
|
"eval_steps_per_second": 0.829, |
|
"step": 924 |
|
}, |
|
{ |
|
"epoch": 132.86, |
|
"learning_rate": 1.8285714285714288e-05, |
|
"loss": 1.4062, |
|
"step": 930 |
|
}, |
|
{ |
|
"epoch": 133.0, |
|
"eval_accuracy": 0.6461538461538462, |
|
"eval_loss": 2.3368117809295654, |
|
"eval_runtime": 2.423, |
|
"eval_samples_per_second": 53.653, |
|
"eval_steps_per_second": 0.825, |
|
"step": 931 |
|
}, |
|
{ |
|
"epoch": 134.0, |
|
"eval_accuracy": 0.6230769230769231, |
|
"eval_loss": 2.3077619075775146, |
|
"eval_runtime": 2.4048, |
|
"eval_samples_per_second": 54.058, |
|
"eval_steps_per_second": 0.832, |
|
"step": 938 |
|
}, |
|
{ |
|
"epoch": 134.29, |
|
"learning_rate": 1.8095238095238097e-05, |
|
"loss": 1.3738, |
|
"step": 940 |
|
}, |
|
{ |
|
"epoch": 135.0, |
|
"eval_accuracy": 0.6307692307692307, |
|
"eval_loss": 2.3112919330596924, |
|
"eval_runtime": 2.391, |
|
"eval_samples_per_second": 54.37, |
|
"eval_steps_per_second": 0.836, |
|
"step": 945 |
|
}, |
|
{ |
|
"epoch": 135.71, |
|
"learning_rate": 1.7904761904761907e-05, |
|
"loss": 1.3566, |
|
"step": 950 |
|
}, |
|
{ |
|
"epoch": 136.0, |
|
"eval_accuracy": 0.6230769230769231, |
|
"eval_loss": 2.2861361503601074, |
|
"eval_runtime": 2.4243, |
|
"eval_samples_per_second": 53.624, |
|
"eval_steps_per_second": 0.825, |
|
"step": 952 |
|
}, |
|
{ |
|
"epoch": 137.0, |
|
"eval_accuracy": 0.6538461538461539, |
|
"eval_loss": 2.2651076316833496, |
|
"eval_runtime": 2.4617, |
|
"eval_samples_per_second": 52.808, |
|
"eval_steps_per_second": 0.812, |
|
"step": 959 |
|
}, |
|
{ |
|
"epoch": 137.14, |
|
"learning_rate": 1.7714285714285717e-05, |
|
"loss": 1.3463, |
|
"step": 960 |
|
}, |
|
{ |
|
"epoch": 138.0, |
|
"eval_accuracy": 0.6615384615384615, |
|
"eval_loss": 2.2631070613861084, |
|
"eval_runtime": 2.4282, |
|
"eval_samples_per_second": 53.537, |
|
"eval_steps_per_second": 0.824, |
|
"step": 966 |
|
}, |
|
{ |
|
"epoch": 138.57, |
|
"learning_rate": 1.7523809523809526e-05, |
|
"loss": 1.3058, |
|
"step": 970 |
|
}, |
|
{ |
|
"epoch": 139.0, |
|
"eval_accuracy": 0.6615384615384615, |
|
"eval_loss": 2.2603201866149902, |
|
"eval_runtime": 2.5878, |
|
"eval_samples_per_second": 50.235, |
|
"eval_steps_per_second": 0.773, |
|
"step": 973 |
|
}, |
|
{ |
|
"epoch": 140.0, |
|
"learning_rate": 1.7333333333333336e-05, |
|
"loss": 1.2967, |
|
"step": 980 |
|
}, |
|
{ |
|
"epoch": 140.0, |
|
"eval_accuracy": 0.6461538461538462, |
|
"eval_loss": 2.2615151405334473, |
|
"eval_runtime": 2.5189, |
|
"eval_samples_per_second": 51.611, |
|
"eval_steps_per_second": 0.794, |
|
"step": 980 |
|
}, |
|
{ |
|
"epoch": 141.0, |
|
"eval_accuracy": 0.6461538461538462, |
|
"eval_loss": 2.218820810317993, |
|
"eval_runtime": 2.4255, |
|
"eval_samples_per_second": 53.597, |
|
"eval_steps_per_second": 0.825, |
|
"step": 987 |
|
}, |
|
{ |
|
"epoch": 141.43, |
|
"learning_rate": 1.7142857142857142e-05, |
|
"loss": 1.2841, |
|
"step": 990 |
|
}, |
|
{ |
|
"epoch": 142.0, |
|
"eval_accuracy": 0.6615384615384615, |
|
"eval_loss": 2.242035388946533, |
|
"eval_runtime": 2.4254, |
|
"eval_samples_per_second": 53.599, |
|
"eval_steps_per_second": 0.825, |
|
"step": 994 |
|
}, |
|
{ |
|
"epoch": 142.86, |
|
"learning_rate": 1.6952380952380955e-05, |
|
"loss": 1.24, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 143.0, |
|
"eval_accuracy": 0.6307692307692307, |
|
"eval_loss": 2.226486921310425, |
|
"eval_runtime": 2.38, |
|
"eval_samples_per_second": 54.621, |
|
"eval_steps_per_second": 0.84, |
|
"step": 1001 |
|
}, |
|
{ |
|
"epoch": 144.0, |
|
"eval_accuracy": 0.6461538461538462, |
|
"eval_loss": 2.212893486022949, |
|
"eval_runtime": 2.5548, |
|
"eval_samples_per_second": 50.885, |
|
"eval_steps_per_second": 0.783, |
|
"step": 1008 |
|
}, |
|
{ |
|
"epoch": 144.29, |
|
"learning_rate": 1.6761904761904764e-05, |
|
"loss": 1.2285, |
|
"step": 1010 |
|
}, |
|
{ |
|
"epoch": 145.0, |
|
"eval_accuracy": 0.6615384615384615, |
|
"eval_loss": 2.227419376373291, |
|
"eval_runtime": 2.4191, |
|
"eval_samples_per_second": 53.738, |
|
"eval_steps_per_second": 0.827, |
|
"step": 1015 |
|
}, |
|
{ |
|
"epoch": 145.71, |
|
"learning_rate": 1.6571428571428574e-05, |
|
"loss": 1.2152, |
|
"step": 1020 |
|
}, |
|
{ |
|
"epoch": 146.0, |
|
"eval_accuracy": 0.6846153846153846, |
|
"eval_loss": 2.18461537361145, |
|
"eval_runtime": 2.34, |
|
"eval_samples_per_second": 55.556, |
|
"eval_steps_per_second": 0.855, |
|
"step": 1022 |
|
}, |
|
{ |
|
"epoch": 147.0, |
|
"eval_accuracy": 0.6615384615384615, |
|
"eval_loss": 2.1721718311309814, |
|
"eval_runtime": 2.5232, |
|
"eval_samples_per_second": 51.522, |
|
"eval_steps_per_second": 0.793, |
|
"step": 1029 |
|
}, |
|
{ |
|
"epoch": 147.14, |
|
"learning_rate": 1.6380952380952384e-05, |
|
"loss": 1.187, |
|
"step": 1030 |
|
}, |
|
{ |
|
"epoch": 148.0, |
|
"eval_accuracy": 0.6846153846153846, |
|
"eval_loss": 2.145458936691284, |
|
"eval_runtime": 2.5242, |
|
"eval_samples_per_second": 51.501, |
|
"eval_steps_per_second": 0.792, |
|
"step": 1036 |
|
}, |
|
{ |
|
"epoch": 148.57, |
|
"learning_rate": 1.6190476190476193e-05, |
|
"loss": 1.1794, |
|
"step": 1040 |
|
}, |
|
{ |
|
"epoch": 149.0, |
|
"eval_accuracy": 0.7153846153846154, |
|
"eval_loss": 2.100327730178833, |
|
"eval_runtime": 2.5331, |
|
"eval_samples_per_second": 51.321, |
|
"eval_steps_per_second": 0.79, |
|
"step": 1043 |
|
}, |
|
{ |
|
"epoch": 150.0, |
|
"learning_rate": 1.6000000000000003e-05, |
|
"loss": 1.1654, |
|
"step": 1050 |
|
}, |
|
{ |
|
"epoch": 150.0, |
|
"eval_accuracy": 0.7, |
|
"eval_loss": 2.1009724140167236, |
|
"eval_runtime": 2.3848, |
|
"eval_samples_per_second": 54.512, |
|
"eval_steps_per_second": 0.839, |
|
"step": 1050 |
|
}, |
|
{ |
|
"epoch": 150.0, |
|
"step": 1050, |
|
"total_flos": 6.047083881086976e+18, |
|
"train_loss": 0.2745589065551758, |
|
"train_runtime": 1386.6967, |
|
"train_samples_per_second": 56.249, |
|
"train_steps_per_second": 0.757 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 1050, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 150, |
|
"save_steps": 500, |
|
"total_flos": 6.047083881086976e+18, |
|
"train_batch_size": 80, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|