| { | |
| "best_metric": 0.9534883720930233, | |
| "best_model_checkpoint": "deit-base-distilled-patch16-224-75-fold5/checkpoint-100", | |
| "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": 0.3488372093023256, | |
| "eval_loss": 0.8041083812713623, | |
| "eval_runtime": 0.6249, | |
| "eval_samples_per_second": 68.814, | |
| "eval_steps_per_second": 3.201, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7209302325581395, | |
| "eval_loss": 0.6042466163635254, | |
| "eval_runtime": 0.607, | |
| "eval_samples_per_second": 70.843, | |
| "eval_steps_per_second": 3.295, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6976744186046512, | |
| "eval_loss": 0.6582099199295044, | |
| "eval_runtime": 0.5929, | |
| "eval_samples_per_second": 72.531, | |
| "eval_steps_per_second": 3.374, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6976744186046512, | |
| "eval_loss": 0.700655996799469, | |
| "eval_runtime": 0.6048, | |
| "eval_samples_per_second": 71.093, | |
| "eval_steps_per_second": 3.307, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 4.375382900238037, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.6827, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6976744186046512, | |
| "eval_loss": 0.5652520656585693, | |
| "eval_runtime": 0.6007, | |
| "eval_samples_per_second": 71.587, | |
| "eval_steps_per_second": 3.33, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.4344584047794342, | |
| "eval_runtime": 0.6065, | |
| "eval_samples_per_second": 70.898, | |
| "eval_steps_per_second": 3.298, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.44925662875175476, | |
| "eval_runtime": 0.6087, | |
| "eval_samples_per_second": 70.643, | |
| "eval_steps_per_second": 3.286, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.47643980383872986, | |
| "eval_runtime": 0.6082, | |
| "eval_samples_per_second": 70.705, | |
| "eval_steps_per_second": 3.289, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.3636239171028137, | |
| "eval_runtime": 0.6061, | |
| "eval_samples_per_second": 70.943, | |
| "eval_steps_per_second": 3.3, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.3277673721313477, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4147, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.2768890857696533, | |
| "eval_runtime": 0.6151, | |
| "eval_samples_per_second": 69.907, | |
| "eval_steps_per_second": 3.252, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.35381999611854553, | |
| "eval_runtime": 0.6067, | |
| "eval_samples_per_second": 70.869, | |
| "eval_steps_per_second": 3.296, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.3320469856262207, | |
| "eval_runtime": 0.605, | |
| "eval_samples_per_second": 71.076, | |
| "eval_steps_per_second": 3.306, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.2667577266693115, | |
| "eval_runtime": 0.6119, | |
| "eval_samples_per_second": 70.272, | |
| "eval_steps_per_second": 3.268, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.28225573897361755, | |
| "eval_runtime": 0.6046, | |
| "eval_samples_per_second": 71.122, | |
| "eval_steps_per_second": 3.308, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.5156199932098389, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.3405, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.48642632365226746, | |
| "eval_runtime": 0.6124, | |
| "eval_samples_per_second": 70.212, | |
| "eval_steps_per_second": 3.266, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.44160977005958557, | |
| "eval_runtime": 0.609, | |
| "eval_samples_per_second": 70.608, | |
| "eval_steps_per_second": 3.284, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.3585019111633301, | |
| "eval_runtime": 0.5963, | |
| "eval_samples_per_second": 72.111, | |
| "eval_steps_per_second": 3.354, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.47688964009284973, | |
| "eval_runtime": 0.6095, | |
| "eval_samples_per_second": 70.545, | |
| "eval_steps_per_second": 3.281, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.5918123722076416, | |
| "eval_runtime": 0.6107, | |
| "eval_samples_per_second": 70.412, | |
| "eval_steps_per_second": 3.275, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 2.0539426803588867, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.3074, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.4116123914718628, | |
| "eval_runtime": 0.6061, | |
| "eval_samples_per_second": 70.94, | |
| "eval_steps_per_second": 3.3, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.4578164219856262, | |
| "eval_runtime": 0.6053, | |
| "eval_samples_per_second": 71.035, | |
| "eval_steps_per_second": 3.304, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.4294234812259674, | |
| "eval_runtime": 0.6208, | |
| "eval_samples_per_second": 69.27, | |
| "eval_steps_per_second": 3.222, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.540077269077301, | |
| "eval_runtime": 0.6113, | |
| "eval_samples_per_second": 70.345, | |
| "eval_steps_per_second": 3.272, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.20692048966884613, | |
| "eval_runtime": 0.6022, | |
| "eval_samples_per_second": 71.403, | |
| "eval_steps_per_second": 3.321, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 5.078647136688232, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.2155, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.22156542539596558, | |
| "eval_runtime": 0.6009, | |
| "eval_samples_per_second": 71.556, | |
| "eval_steps_per_second": 3.328, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.38367632031440735, | |
| "eval_runtime": 0.6007, | |
| "eval_samples_per_second": 71.58, | |
| "eval_steps_per_second": 3.329, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.24731607735157013, | |
| "eval_runtime": 0.6066, | |
| "eval_samples_per_second": 70.882, | |
| "eval_steps_per_second": 3.297, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.22330066561698914, | |
| "eval_runtime": 0.612, | |
| "eval_samples_per_second": 70.259, | |
| "eval_steps_per_second": 3.268, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3489048182964325, | |
| "eval_runtime": 0.6096, | |
| "eval_samples_per_second": 70.543, | |
| "eval_steps_per_second": 3.281, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 3.3193013668060303, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.1874, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.4634832739830017, | |
| "eval_runtime": 0.6073, | |
| "eval_samples_per_second": 70.802, | |
| "eval_steps_per_second": 3.293, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.24913936853408813, | |
| "eval_runtime": 0.6169, | |
| "eval_samples_per_second": 69.704, | |
| "eval_steps_per_second": 3.242, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.2564470171928406, | |
| "eval_runtime": 0.6188, | |
| "eval_samples_per_second": 69.484, | |
| "eval_steps_per_second": 3.232, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3655398488044739, | |
| "eval_runtime": 0.6099, | |
| "eval_samples_per_second": 70.504, | |
| "eval_steps_per_second": 3.279, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.24981090426445007, | |
| "eval_runtime": 0.6235, | |
| "eval_samples_per_second": 68.962, | |
| "eval_steps_per_second": 3.208, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "grad_norm": 1.9860986471176147, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.1324, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.2921779453754425, | |
| "eval_runtime": 0.6126, | |
| "eval_samples_per_second": 70.191, | |
| "eval_steps_per_second": 3.265, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.5368495583534241, | |
| "eval_runtime": 0.6089, | |
| "eval_samples_per_second": 70.616, | |
| "eval_steps_per_second": 3.284, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.5738598108291626, | |
| "eval_runtime": 0.6131, | |
| "eval_samples_per_second": 70.13, | |
| "eval_steps_per_second": 3.262, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.5049097537994385, | |
| "eval_runtime": 0.6178, | |
| "eval_samples_per_second": 69.6, | |
| "eval_steps_per_second": 3.237, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.5902883410453796, | |
| "eval_runtime": 0.6127, | |
| "eval_samples_per_second": 70.176, | |
| "eval_steps_per_second": 3.264, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 2.2311851978302, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.1222, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.4885583817958832, | |
| "eval_runtime": 0.6261, | |
| "eval_samples_per_second": 68.678, | |
| "eval_steps_per_second": 3.194, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.41742998361587524, | |
| "eval_runtime": 0.6335, | |
| "eval_samples_per_second": 67.877, | |
| "eval_steps_per_second": 3.157, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.542888343334198, | |
| "eval_runtime": 0.5996, | |
| "eval_samples_per_second": 71.718, | |
| "eval_steps_per_second": 3.336, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.6896758675575256, | |
| "eval_runtime": 0.6147, | |
| "eval_samples_per_second": 69.953, | |
| "eval_steps_per_second": 3.254, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.6804940700531006, | |
| "eval_runtime": 0.615, | |
| "eval_samples_per_second": 69.918, | |
| "eval_steps_per_second": 3.252, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "grad_norm": 2.4388444423675537, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.1008, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.4073416590690613, | |
| "eval_runtime": 0.6191, | |
| "eval_samples_per_second": 69.46, | |
| "eval_steps_per_second": 3.231, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.4160747230052948, | |
| "eval_runtime": 0.6162, | |
| "eval_samples_per_second": 69.777, | |
| "eval_steps_per_second": 3.245, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.6485046744346619, | |
| "eval_runtime": 0.6132, | |
| "eval_samples_per_second": 70.12, | |
| "eval_steps_per_second": 3.261, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.6746022701263428, | |
| "eval_runtime": 0.6175, | |
| "eval_samples_per_second": 69.639, | |
| "eval_steps_per_second": 3.239, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.44333821535110474, | |
| "eval_runtime": 0.6198, | |
| "eval_samples_per_second": 69.374, | |
| "eval_steps_per_second": 3.227, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 1.96018385887146, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.117, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.9534883720930233, | |
| "eval_loss": 0.278799444437027, | |
| "eval_runtime": 0.6169, | |
| "eval_samples_per_second": 69.706, | |
| "eval_steps_per_second": 3.242, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3440650999546051, | |
| "eval_runtime": 0.6161, | |
| "eval_samples_per_second": 69.797, | |
| "eval_steps_per_second": 3.246, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.46627554297447205, | |
| "eval_runtime": 0.6164, | |
| "eval_samples_per_second": 69.765, | |
| "eval_steps_per_second": 3.245, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.3299558460712433, | |
| "eval_runtime": 0.6159, | |
| "eval_samples_per_second": 69.814, | |
| "eval_steps_per_second": 3.247, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_accuracy": 0.9534883720930233, | |
| "eval_loss": 0.25306156277656555, | |
| "eval_runtime": 0.6221, | |
| "eval_samples_per_second": 69.12, | |
| "eval_steps_per_second": 3.215, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "grad_norm": 1.9721273183822632, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0879, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "eval_accuracy": 0.9534883720930233, | |
| "eval_loss": 0.22605039179325104, | |
| "eval_runtime": 0.616, | |
| "eval_samples_per_second": 69.808, | |
| "eval_steps_per_second": 3.247, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.34897878766059875, | |
| "eval_runtime": 0.6198, | |
| "eval_samples_per_second": 69.373, | |
| "eval_steps_per_second": 3.227, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.5059694647789001, | |
| "eval_runtime": 0.6134, | |
| "eval_samples_per_second": 70.104, | |
| "eval_steps_per_second": 3.261, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.4846828877925873, | |
| "eval_runtime": 0.6108, | |
| "eval_samples_per_second": 70.402, | |
| "eval_steps_per_second": 3.274, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3603507876396179, | |
| "eval_runtime": 0.6084, | |
| "eval_samples_per_second": 70.675, | |
| "eval_steps_per_second": 3.287, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 5.5369553565979, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.0898, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3626036047935486, | |
| "eval_runtime": 0.6067, | |
| "eval_samples_per_second": 70.87, | |
| "eval_steps_per_second": 3.296, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.4641226530075073, | |
| "eval_runtime": 0.6132, | |
| "eval_samples_per_second": 70.124, | |
| "eval_steps_per_second": 3.262, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.4605565667152405, | |
| "eval_runtime": 0.6118, | |
| "eval_samples_per_second": 70.282, | |
| "eval_steps_per_second": 3.269, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.34193670749664307, | |
| "eval_runtime": 0.6144, | |
| "eval_samples_per_second": 69.988, | |
| "eval_steps_per_second": 3.255, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.3331243693828583, | |
| "eval_runtime": 0.6254, | |
| "eval_samples_per_second": 68.751, | |
| "eval_steps_per_second": 3.198, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "grad_norm": 2.0372941493988037, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.095, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3803730010986328, | |
| "eval_runtime": 0.6294, | |
| "eval_samples_per_second": 68.319, | |
| "eval_steps_per_second": 3.178, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.37833139300346375, | |
| "eval_runtime": 0.6275, | |
| "eval_samples_per_second": 68.521, | |
| "eval_steps_per_second": 3.187, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3523567020893097, | |
| "eval_runtime": 0.6243, | |
| "eval_samples_per_second": 68.873, | |
| "eval_steps_per_second": 3.203, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.423170804977417, | |
| "eval_runtime": 0.675, | |
| "eval_samples_per_second": 63.699, | |
| "eval_steps_per_second": 2.963, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.41566768288612366, | |
| "eval_runtime": 0.6177, | |
| "eval_samples_per_second": 69.612, | |
| "eval_steps_per_second": 3.238, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "grad_norm": 1.394300103187561, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.0735, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.493731826543808, | |
| "eval_runtime": 0.615, | |
| "eval_samples_per_second": 69.919, | |
| "eval_steps_per_second": 3.252, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 71.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.4616422951221466, | |
| "eval_runtime": 0.6152, | |
| "eval_samples_per_second": 69.892, | |
| "eval_steps_per_second": 3.251, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.43878328800201416, | |
| "eval_runtime": 0.615, | |
| "eval_samples_per_second": 69.921, | |
| "eval_steps_per_second": 3.252, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.38112929463386536, | |
| "eval_runtime": 0.6155, | |
| "eval_samples_per_second": 69.861, | |
| "eval_steps_per_second": 3.249, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.25220033526420593, | |
| "eval_runtime": 0.61, | |
| "eval_samples_per_second": 70.495, | |
| "eval_steps_per_second": 3.279, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "grad_norm": 1.9808557033538818, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.0855, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "eval_accuracy": 0.9534883720930233, | |
| "eval_loss": 0.16766749322414398, | |
| "eval_runtime": 0.6139, | |
| "eval_samples_per_second": 70.05, | |
| "eval_steps_per_second": 3.258, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.9534883720930233, | |
| "eval_loss": 0.16528721153736115, | |
| "eval_runtime": 0.6142, | |
| "eval_samples_per_second": 70.01, | |
| "eval_steps_per_second": 3.256, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 77.0, | |
| "eval_accuracy": 0.9534883720930233, | |
| "eval_loss": 0.21727260947227478, | |
| "eval_runtime": 0.6179, | |
| "eval_samples_per_second": 69.594, | |
| "eval_steps_per_second": 3.237, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3596310019493103, | |
| "eval_runtime": 0.6118, | |
| "eval_samples_per_second": 70.286, | |
| "eval_steps_per_second": 3.269, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 79.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.4653131663799286, | |
| "eval_runtime": 0.6097, | |
| "eval_samples_per_second": 70.526, | |
| "eval_steps_per_second": 3.28, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 2.2111823558807373, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.077, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.47392675280570984, | |
| "eval_runtime": 0.6183, | |
| "eval_samples_per_second": 69.54, | |
| "eval_steps_per_second": 3.234, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 81.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.360915869474411, | |
| "eval_runtime": 0.6117, | |
| "eval_samples_per_second": 70.296, | |
| "eval_steps_per_second": 3.27, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_accuracy": 0.9534883720930233, | |
| "eval_loss": 0.23335148394107819, | |
| "eval_runtime": 0.6143, | |
| "eval_samples_per_second": 70.004, | |
| "eval_steps_per_second": 3.256, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 83.0, | |
| "eval_accuracy": 0.9534883720930233, | |
| "eval_loss": 0.21253377199172974, | |
| "eval_runtime": 0.6169, | |
| "eval_samples_per_second": 69.706, | |
| "eval_steps_per_second": 3.242, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 0.9534883720930233, | |
| "eval_loss": 0.2594672441482544, | |
| "eval_runtime": 0.6281, | |
| "eval_samples_per_second": 68.465, | |
| "eval_steps_per_second": 3.184, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "grad_norm": 1.4013921022415161, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.0731, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.37020593881607056, | |
| "eval_runtime": 0.6591, | |
| "eval_samples_per_second": 65.245, | |
| "eval_steps_per_second": 3.035, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.4634316563606262, | |
| "eval_runtime": 0.6372, | |
| "eval_samples_per_second": 67.483, | |
| "eval_steps_per_second": 3.139, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 87.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.5253772139549255, | |
| "eval_runtime": 0.6706, | |
| "eval_samples_per_second": 64.119, | |
| "eval_steps_per_second": 2.982, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.5813474655151367, | |
| "eval_runtime": 0.6652, | |
| "eval_samples_per_second": 64.637, | |
| "eval_steps_per_second": 3.006, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 89.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.5628194212913513, | |
| "eval_runtime": 0.6372, | |
| "eval_samples_per_second": 67.487, | |
| "eval_steps_per_second": 3.139, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "grad_norm": 1.3443642854690552, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.078, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.5479036569595337, | |
| "eval_runtime": 0.6115, | |
| "eval_samples_per_second": 70.317, | |
| "eval_steps_per_second": 3.271, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 91.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.5215557217597961, | |
| "eval_runtime": 0.6168, | |
| "eval_samples_per_second": 69.718, | |
| "eval_steps_per_second": 3.243, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.509810745716095, | |
| "eval_runtime": 0.6149, | |
| "eval_samples_per_second": 69.933, | |
| "eval_steps_per_second": 3.253, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 93.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.5243856310844421, | |
| "eval_runtime": 0.6268, | |
| "eval_samples_per_second": 68.603, | |
| "eval_steps_per_second": 3.191, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 94.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.5471506118774414, | |
| "eval_runtime": 0.622, | |
| "eval_samples_per_second": 69.132, | |
| "eval_steps_per_second": 3.215, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "grad_norm": 1.141045331954956, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.079, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.558978259563446, | |
| "eval_runtime": 0.6529, | |
| "eval_samples_per_second": 65.862, | |
| "eval_steps_per_second": 3.063, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.5589279532432556, | |
| "eval_runtime": 0.6301, | |
| "eval_samples_per_second": 68.238, | |
| "eval_steps_per_second": 3.174, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 97.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.5455338954925537, | |
| "eval_runtime": 0.6365, | |
| "eval_samples_per_second": 67.555, | |
| "eval_steps_per_second": 3.142, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 98.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.5321924686431885, | |
| "eval_runtime": 0.62, | |
| "eval_samples_per_second": 69.358, | |
| "eval_steps_per_second": 3.226, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.5214679837226868, | |
| "eval_runtime": 0.6383, | |
| "eval_samples_per_second": 67.365, | |
| "eval_steps_per_second": 3.133, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "grad_norm": 1.9139795303344727, | |
| "learning_rate": 0.0, | |
| "loss": 0.0477, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.5201581716537476, | |
| "eval_runtime": 0.624, | |
| "eval_samples_per_second": 68.908, | |
| "eval_steps_per_second": 3.205, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 200, | |
| "total_flos": 1.867590382823424e+18, | |
| "train_loss": 0.170336701720953, | |
| "train_runtime": 1121.821, | |
| "train_samples_per_second": 21.483, | |
| "train_steps_per_second": 0.178 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.9534883720930233, | |
| "eval_loss": 0.278799444437027, | |
| "eval_runtime": 0.6419, | |
| "eval_samples_per_second": 66.992, | |
| "eval_steps_per_second": 3.116, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.867590382823424e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |