| { | |
| "best_metric": 0.9534883720930233, | |
| "best_model_checkpoint": "beit-base-patch16-224-75-fold2/checkpoint-40", | |
| "epoch": 100.0, | |
| "eval_steps": 500, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5348837209302325, | |
| "eval_loss": 0.7091274261474609, | |
| "eval_runtime": 0.6801, | |
| "eval_samples_per_second": 63.228, | |
| "eval_steps_per_second": 2.941, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7209302325581395, | |
| "eval_loss": 0.6502025723457336, | |
| "eval_runtime": 0.7085, | |
| "eval_samples_per_second": 60.688, | |
| "eval_steps_per_second": 2.823, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6976744186046512, | |
| "eval_loss": 0.9193121194839478, | |
| "eval_runtime": 0.6883, | |
| "eval_samples_per_second": 62.471, | |
| "eval_steps_per_second": 2.906, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7441860465116279, | |
| "eval_loss": 0.7498829960823059, | |
| "eval_runtime": 0.707, | |
| "eval_samples_per_second": 60.819, | |
| "eval_steps_per_second": 2.829, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 9.87000846862793, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.6436, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.45272693037986755, | |
| "eval_runtime": 0.6566, | |
| "eval_samples_per_second": 65.484, | |
| "eval_steps_per_second": 3.046, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.4168838560581207, | |
| "eval_runtime": 0.6591, | |
| "eval_samples_per_second": 65.241, | |
| "eval_steps_per_second": 3.034, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7441860465116279, | |
| "eval_loss": 0.5772562026977539, | |
| "eval_runtime": 0.6565, | |
| "eval_samples_per_second": 65.498, | |
| "eval_steps_per_second": 3.046, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.4076061546802521, | |
| "eval_runtime": 0.6649, | |
| "eval_samples_per_second": 64.672, | |
| "eval_steps_per_second": 3.008, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.39387083053588867, | |
| "eval_runtime": 0.6727, | |
| "eval_samples_per_second": 63.921, | |
| "eval_steps_per_second": 2.973, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 8.721357345581055, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3863, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.40172407031059265, | |
| "eval_runtime": 0.6905, | |
| "eval_samples_per_second": 62.273, | |
| "eval_steps_per_second": 2.896, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.4918433725833893, | |
| "eval_runtime": 0.6742, | |
| "eval_samples_per_second": 63.784, | |
| "eval_steps_per_second": 2.967, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.26880943775177, | |
| "eval_runtime": 0.6795, | |
| "eval_samples_per_second": 63.277, | |
| "eval_steps_per_second": 2.943, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.38842692971229553, | |
| "eval_runtime": 0.6685, | |
| "eval_samples_per_second": 64.32, | |
| "eval_steps_per_second": 2.992, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.36793774366378784, | |
| "eval_runtime": 0.6633, | |
| "eval_samples_per_second": 64.825, | |
| "eval_steps_per_second": 3.015, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 7.428971767425537, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.2925, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.28020721673965454, | |
| "eval_runtime": 0.6621, | |
| "eval_samples_per_second": 64.946, | |
| "eval_steps_per_second": 3.021, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.2435620129108429, | |
| "eval_runtime": 0.6853, | |
| "eval_samples_per_second": 62.742, | |
| "eval_steps_per_second": 2.918, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.23368221521377563, | |
| "eval_runtime": 0.6637, | |
| "eval_samples_per_second": 64.786, | |
| "eval_steps_per_second": 3.013, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.3710887134075165, | |
| "eval_runtime": 0.6636, | |
| "eval_samples_per_second": 64.797, | |
| "eval_steps_per_second": 3.014, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.23721440136432648, | |
| "eval_runtime": 0.6651, | |
| "eval_samples_per_second": 64.653, | |
| "eval_steps_per_second": 3.007, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 6.218175888061523, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.2289, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9534883720930233, | |
| "eval_loss": 0.2685289978981018, | |
| "eval_runtime": 0.6636, | |
| "eval_samples_per_second": 64.799, | |
| "eval_steps_per_second": 3.014, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.26095083355903625, | |
| "eval_runtime": 0.6603, | |
| "eval_samples_per_second": 65.125, | |
| "eval_steps_per_second": 3.029, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.33278653025627136, | |
| "eval_runtime": 0.6615, | |
| "eval_samples_per_second": 65.007, | |
| "eval_steps_per_second": 3.024, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.34786635637283325, | |
| "eval_runtime": 0.6666, | |
| "eval_samples_per_second": 64.504, | |
| "eval_steps_per_second": 3.0, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.28547996282577515, | |
| "eval_runtime": 0.6589, | |
| "eval_samples_per_second": 65.256, | |
| "eval_steps_per_second": 3.035, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 2.4903624057769775, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.219, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.29620176553726196, | |
| "eval_runtime": 0.7126, | |
| "eval_samples_per_second": 60.347, | |
| "eval_steps_per_second": 2.807, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.4037957191467285, | |
| "eval_runtime": 0.664, | |
| "eval_samples_per_second": 64.756, | |
| "eval_steps_per_second": 3.012, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.31493234634399414, | |
| "eval_runtime": 0.6564, | |
| "eval_samples_per_second": 65.513, | |
| "eval_steps_per_second": 3.047, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.3212130665779114, | |
| "eval_runtime": 0.6626, | |
| "eval_samples_per_second": 64.896, | |
| "eval_steps_per_second": 3.018, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.48950737714767456, | |
| "eval_runtime": 0.6802, | |
| "eval_samples_per_second": 63.216, | |
| "eval_steps_per_second": 2.94, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 11.771470069885254, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.1933, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.4335302412509918, | |
| "eval_runtime": 0.6571, | |
| "eval_samples_per_second": 65.442, | |
| "eval_steps_per_second": 3.044, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.3521455228328705, | |
| "eval_runtime": 0.6638, | |
| "eval_samples_per_second": 64.774, | |
| "eval_steps_per_second": 3.013, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.29595664143562317, | |
| "eval_runtime": 0.6692, | |
| "eval_samples_per_second": 64.258, | |
| "eval_steps_per_second": 2.989, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.4036531150341034, | |
| "eval_runtime": 0.6662, | |
| "eval_samples_per_second": 64.546, | |
| "eval_steps_per_second": 3.002, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.2912556529045105, | |
| "eval_runtime": 0.6652, | |
| "eval_samples_per_second": 64.644, | |
| "eval_steps_per_second": 3.007, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "grad_norm": 5.611938953399658, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.1892, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.30434587597846985, | |
| "eval_runtime": 0.6558, | |
| "eval_samples_per_second": 65.57, | |
| "eval_steps_per_second": 3.05, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.3602141737937927, | |
| "eval_runtime": 0.6584, | |
| "eval_samples_per_second": 65.311, | |
| "eval_steps_per_second": 3.038, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.33154794573783875, | |
| "eval_runtime": 0.6651, | |
| "eval_samples_per_second": 64.655, | |
| "eval_steps_per_second": 3.007, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.26741957664489746, | |
| "eval_runtime": 0.6641, | |
| "eval_samples_per_second": 64.748, | |
| "eval_steps_per_second": 3.012, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_accuracy": 0.9534883720930233, | |
| "eval_loss": 0.29697251319885254, | |
| "eval_runtime": 0.6727, | |
| "eval_samples_per_second": 63.918, | |
| "eval_steps_per_second": 2.973, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 3.212876796722412, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.15, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.9534883720930233, | |
| "eval_loss": 0.26611432433128357, | |
| "eval_runtime": 0.6735, | |
| "eval_samples_per_second": 63.847, | |
| "eval_steps_per_second": 2.97, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.2550868093967438, | |
| "eval_runtime": 0.6643, | |
| "eval_samples_per_second": 64.733, | |
| "eval_steps_per_second": 3.011, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.24666233360767365, | |
| "eval_runtime": 0.684, | |
| "eval_samples_per_second": 62.867, | |
| "eval_steps_per_second": 2.924, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_accuracy": 0.9534883720930233, | |
| "eval_loss": 0.3008131682872772, | |
| "eval_runtime": 0.6612, | |
| "eval_samples_per_second": 65.036, | |
| "eval_steps_per_second": 3.025, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.3265131115913391, | |
| "eval_runtime": 0.6639, | |
| "eval_samples_per_second": 64.767, | |
| "eval_steps_per_second": 3.012, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "grad_norm": 3.951080560684204, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.1238, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.2668008506298065, | |
| "eval_runtime": 0.6626, | |
| "eval_samples_per_second": 64.893, | |
| "eval_steps_per_second": 3.018, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.25739309191703796, | |
| "eval_runtime": 0.6671, | |
| "eval_samples_per_second": 64.463, | |
| "eval_steps_per_second": 2.998, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_accuracy": 0.9534883720930233, | |
| "eval_loss": 0.24979150295257568, | |
| "eval_runtime": 0.6578, | |
| "eval_samples_per_second": 65.37, | |
| "eval_steps_per_second": 3.04, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.331857293844223, | |
| "eval_runtime": 0.6644, | |
| "eval_samples_per_second": 64.721, | |
| "eval_steps_per_second": 3.01, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.23575438559055328, | |
| "eval_runtime": 0.6658, | |
| "eval_samples_per_second": 64.585, | |
| "eval_steps_per_second": 3.004, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 4.541098594665527, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.1063, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.2014588862657547, | |
| "eval_runtime": 0.6601, | |
| "eval_samples_per_second": 65.142, | |
| "eval_steps_per_second": 3.03, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.21708515286445618, | |
| "eval_runtime": 0.6871, | |
| "eval_samples_per_second": 62.582, | |
| "eval_steps_per_second": 2.911, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.31185880303382874, | |
| "eval_runtime": 0.6611, | |
| "eval_samples_per_second": 65.04, | |
| "eval_steps_per_second": 3.025, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.26736676692962646, | |
| "eval_runtime": 0.6628, | |
| "eval_samples_per_second": 64.878, | |
| "eval_steps_per_second": 3.018, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.307630330324173, | |
| "eval_runtime": 0.6615, | |
| "eval_samples_per_second": 65.008, | |
| "eval_steps_per_second": 3.024, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "grad_norm": 5.14859676361084, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1112, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3181542158126831, | |
| "eval_runtime": 0.6603, | |
| "eval_samples_per_second": 65.12, | |
| "eval_steps_per_second": 3.029, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.3371391296386719, | |
| "eval_runtime": 0.662, | |
| "eval_samples_per_second": 64.957, | |
| "eval_steps_per_second": 3.021, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.35395121574401855, | |
| "eval_runtime": 0.6647, | |
| "eval_samples_per_second": 64.692, | |
| "eval_steps_per_second": 3.009, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.40581977367401123, | |
| "eval_runtime": 0.6636, | |
| "eval_samples_per_second": 64.798, | |
| "eval_steps_per_second": 3.014, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.4012940526008606, | |
| "eval_runtime": 0.6602, | |
| "eval_samples_per_second": 65.133, | |
| "eval_steps_per_second": 3.029, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 5.5886430740356445, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.1128, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.3309071362018585, | |
| "eval_runtime": 0.662, | |
| "eval_samples_per_second": 64.952, | |
| "eval_steps_per_second": 3.021, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.3272179365158081, | |
| "eval_runtime": 0.6657, | |
| "eval_samples_per_second": 64.595, | |
| "eval_steps_per_second": 3.004, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.401218980550766, | |
| "eval_runtime": 0.6635, | |
| "eval_samples_per_second": 64.804, | |
| "eval_steps_per_second": 3.014, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.5794376730918884, | |
| "eval_runtime": 0.6788, | |
| "eval_samples_per_second": 63.351, | |
| "eval_steps_per_second": 2.947, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.3881092369556427, | |
| "eval_runtime": 0.6632, | |
| "eval_samples_per_second": 64.835, | |
| "eval_steps_per_second": 3.016, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "grad_norm": 3.247230291366577, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.1168, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.2990245819091797, | |
| "eval_runtime": 0.6569, | |
| "eval_samples_per_second": 65.462, | |
| "eval_steps_per_second": 3.045, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3017998933792114, | |
| "eval_runtime": 0.666, | |
| "eval_samples_per_second": 64.564, | |
| "eval_steps_per_second": 3.003, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.2560558319091797, | |
| "eval_runtime": 0.6618, | |
| "eval_samples_per_second": 64.974, | |
| "eval_steps_per_second": 3.022, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.2920568883419037, | |
| "eval_runtime": 0.6666, | |
| "eval_samples_per_second": 64.509, | |
| "eval_steps_per_second": 3.0, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.325835257768631, | |
| "eval_runtime": 0.6696, | |
| "eval_samples_per_second": 64.221, | |
| "eval_steps_per_second": 2.987, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "grad_norm": 4.725736618041992, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.0846, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.29250141978263855, | |
| "eval_runtime": 0.6576, | |
| "eval_samples_per_second": 65.394, | |
| "eval_steps_per_second": 3.042, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 71.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.3073433041572571, | |
| "eval_runtime": 0.6724, | |
| "eval_samples_per_second": 63.946, | |
| "eval_steps_per_second": 2.974, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.33179721236228943, | |
| "eval_runtime": 0.6603, | |
| "eval_samples_per_second": 65.125, | |
| "eval_steps_per_second": 3.029, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.3426916003227234, | |
| "eval_runtime": 0.6637, | |
| "eval_samples_per_second": 64.785, | |
| "eval_steps_per_second": 3.013, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.3587954342365265, | |
| "eval_runtime": 0.6974, | |
| "eval_samples_per_second": 61.655, | |
| "eval_steps_per_second": 2.868, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "grad_norm": 2.8244574069976807, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.0845, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.393873006105423, | |
| "eval_runtime": 0.6553, | |
| "eval_samples_per_second": 65.622, | |
| "eval_steps_per_second": 3.052, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.37742680311203003, | |
| "eval_runtime": 0.6587, | |
| "eval_samples_per_second": 65.284, | |
| "eval_steps_per_second": 3.036, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 77.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.37461453676223755, | |
| "eval_runtime": 0.6678, | |
| "eval_samples_per_second": 64.39, | |
| "eval_steps_per_second": 2.995, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.4072588086128235, | |
| "eval_runtime": 0.6788, | |
| "eval_samples_per_second": 63.346, | |
| "eval_steps_per_second": 2.946, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 79.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.388566255569458, | |
| "eval_runtime": 0.6641, | |
| "eval_samples_per_second": 64.748, | |
| "eval_steps_per_second": 3.012, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 3.346651077270508, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.0885, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.37651050090789795, | |
| "eval_runtime": 0.6646, | |
| "eval_samples_per_second": 64.705, | |
| "eval_steps_per_second": 3.01, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 81.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.3976840674877167, | |
| "eval_runtime": 0.669, | |
| "eval_samples_per_second": 64.274, | |
| "eval_steps_per_second": 2.989, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.3863917887210846, | |
| "eval_runtime": 0.7197, | |
| "eval_samples_per_second": 59.75, | |
| "eval_steps_per_second": 2.779, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 83.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.3809366524219513, | |
| "eval_runtime": 0.6665, | |
| "eval_samples_per_second": 64.517, | |
| "eval_steps_per_second": 3.001, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.44918695092201233, | |
| "eval_runtime": 0.6947, | |
| "eval_samples_per_second": 61.893, | |
| "eval_steps_per_second": 2.879, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "grad_norm": 5.120132923126221, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.0859, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.5478556156158447, | |
| "eval_runtime": 0.6681, | |
| "eval_samples_per_second": 64.36, | |
| "eval_steps_per_second": 2.994, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.5372174978256226, | |
| "eval_runtime": 0.6829, | |
| "eval_samples_per_second": 62.964, | |
| "eval_steps_per_second": 2.929, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 87.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.45123860239982605, | |
| "eval_runtime": 0.6755, | |
| "eval_samples_per_second": 63.653, | |
| "eval_steps_per_second": 2.961, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.392961323261261, | |
| "eval_runtime": 0.6724, | |
| "eval_samples_per_second": 63.95, | |
| "eval_steps_per_second": 2.974, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 89.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.3842366635799408, | |
| "eval_runtime": 0.6851, | |
| "eval_samples_per_second": 62.769, | |
| "eval_steps_per_second": 2.919, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "grad_norm": 3.358522415161133, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0764, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.3807728886604309, | |
| "eval_runtime": 0.674, | |
| "eval_samples_per_second": 63.803, | |
| "eval_steps_per_second": 2.968, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 91.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.37874656915664673, | |
| "eval_runtime": 0.675, | |
| "eval_samples_per_second": 63.702, | |
| "eval_steps_per_second": 2.963, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.38331902027130127, | |
| "eval_runtime": 0.6676, | |
| "eval_samples_per_second": 64.412, | |
| "eval_steps_per_second": 2.996, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 93.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.3912203907966614, | |
| "eval_runtime": 0.6718, | |
| "eval_samples_per_second": 64.012, | |
| "eval_steps_per_second": 2.977, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 94.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3887650668621063, | |
| "eval_runtime": 0.6768, | |
| "eval_samples_per_second": 63.537, | |
| "eval_steps_per_second": 2.955, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "grad_norm": 1.5960582494735718, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.0727, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3817383944988251, | |
| "eval_runtime": 0.6624, | |
| "eval_samples_per_second": 64.917, | |
| "eval_steps_per_second": 3.019, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.3707771599292755, | |
| "eval_runtime": 0.6663, | |
| "eval_samples_per_second": 64.531, | |
| "eval_steps_per_second": 3.001, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 97.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.3639983832836151, | |
| "eval_runtime": 0.6611, | |
| "eval_samples_per_second": 65.04, | |
| "eval_steps_per_second": 3.025, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 98.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.36134788393974304, | |
| "eval_runtime": 0.6704, | |
| "eval_samples_per_second": 64.143, | |
| "eval_steps_per_second": 2.983, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.3606959581375122, | |
| "eval_runtime": 0.6758, | |
| "eval_samples_per_second": 63.632, | |
| "eval_steps_per_second": 2.96, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "grad_norm": 3.710935592651367, | |
| "learning_rate": 0.0, | |
| "loss": 0.069, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.3605453670024872, | |
| "eval_runtime": 0.6509, | |
| "eval_samples_per_second": 66.067, | |
| "eval_steps_per_second": 3.073, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 200, | |
| "total_flos": 1.8667587341684736e+18, | |
| "train_loss": 0.17176154434680937, | |
| "train_runtime": 1200.5875, | |
| "train_samples_per_second": 20.074, | |
| "train_steps_per_second": 0.167 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.9534883720930233, | |
| "eval_loss": 0.2685289978981018, | |
| "eval_runtime": 0.6895, | |
| "eval_samples_per_second": 62.366, | |
| "eval_steps_per_second": 2.901, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 500, | |
| "total_flos": 1.8667587341684736e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |