| { | |
| "best_metric": 0.9302325581395349, | |
| "best_model_checkpoint": "deit-base-distilled-patch16-224-75-fold2/checkpoint-42", | |
| "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": 0.4418604651162791, | |
| "eval_loss": 0.8814871311187744, | |
| "eval_runtime": 0.6043, | |
| "eval_samples_per_second": 71.155, | |
| "eval_steps_per_second": 3.31, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6976744186046512, | |
| "eval_loss": 0.6436183452606201, | |
| "eval_runtime": 0.5962, | |
| "eval_samples_per_second": 72.129, | |
| "eval_steps_per_second": 3.355, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6976744186046512, | |
| "eval_loss": 0.8487775921821594, | |
| "eval_runtime": 0.5985, | |
| "eval_samples_per_second": 71.842, | |
| "eval_steps_per_second": 3.341, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6976744186046512, | |
| "eval_loss": 0.8219477534294128, | |
| "eval_runtime": 0.6021, | |
| "eval_samples_per_second": 71.414, | |
| "eval_steps_per_second": 3.322, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.587785005569458, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.6918, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6976744186046512, | |
| "eval_loss": 0.5491273403167725, | |
| "eval_runtime": 0.6041, | |
| "eval_samples_per_second": 71.178, | |
| "eval_steps_per_second": 3.311, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7209302325581395, | |
| "eval_loss": 0.46033531427383423, | |
| "eval_runtime": 0.6098, | |
| "eval_samples_per_second": 70.52, | |
| "eval_steps_per_second": 3.28, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7441860465116279, | |
| "eval_loss": 0.5602250099182129, | |
| "eval_runtime": 0.6092, | |
| "eval_samples_per_second": 70.59, | |
| "eval_steps_per_second": 3.283, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7441860465116279, | |
| "eval_loss": 0.5694101452827454, | |
| "eval_runtime": 0.6084, | |
| "eval_samples_per_second": 70.681, | |
| "eval_steps_per_second": 3.287, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.4429534077644348, | |
| "eval_runtime": 0.6033, | |
| "eval_samples_per_second": 71.278, | |
| "eval_steps_per_second": 3.315, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 4.65231990814209, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3867, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.38796064257621765, | |
| "eval_runtime": 0.6108, | |
| "eval_samples_per_second": 70.404, | |
| "eval_steps_per_second": 3.275, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.5069238543510437, | |
| "eval_runtime": 0.6141, | |
| "eval_samples_per_second": 70.023, | |
| "eval_steps_per_second": 3.257, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.3738561272621155, | |
| "eval_runtime": 0.6236, | |
| "eval_samples_per_second": 68.956, | |
| "eval_steps_per_second": 3.207, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.2981454133987427, | |
| "eval_runtime": 0.6222, | |
| "eval_samples_per_second": 69.104, | |
| "eval_steps_per_second": 3.214, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.31695839762687683, | |
| "eval_runtime": 0.6332, | |
| "eval_samples_per_second": 67.904, | |
| "eval_steps_per_second": 3.158, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 3.3840270042419434, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.2722, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.251080721616745, | |
| "eval_runtime": 0.6281, | |
| "eval_samples_per_second": 68.457, | |
| "eval_steps_per_second": 3.184, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.24079930782318115, | |
| "eval_runtime": 0.6102, | |
| "eval_samples_per_second": 70.465, | |
| "eval_steps_per_second": 3.277, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.37512481212615967, | |
| "eval_runtime": 0.6106, | |
| "eval_samples_per_second": 70.421, | |
| "eval_steps_per_second": 3.275, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.30806484818458557, | |
| "eval_runtime": 0.6459, | |
| "eval_samples_per_second": 66.573, | |
| "eval_steps_per_second": 3.096, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.24885986745357513, | |
| "eval_runtime": 0.6138, | |
| "eval_samples_per_second": 70.053, | |
| "eval_steps_per_second": 3.258, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 3.2458279132843018, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.209, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.2802305817604065, | |
| "eval_runtime": 0.6134, | |
| "eval_samples_per_second": 70.101, | |
| "eval_steps_per_second": 3.261, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.26246392726898193, | |
| "eval_runtime": 0.6143, | |
| "eval_samples_per_second": 70.001, | |
| "eval_steps_per_second": 3.256, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.25953027606010437, | |
| "eval_runtime": 0.6159, | |
| "eval_samples_per_second": 69.812, | |
| "eval_steps_per_second": 3.247, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.504776120185852, | |
| "eval_runtime": 0.6212, | |
| "eval_samples_per_second": 69.226, | |
| "eval_steps_per_second": 3.22, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.28795742988586426, | |
| "eval_runtime": 0.6339, | |
| "eval_samples_per_second": 67.832, | |
| "eval_steps_per_second": 3.155, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 3.2135868072509766, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.2027, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.28604379296302795, | |
| "eval_runtime": 0.6257, | |
| "eval_samples_per_second": 68.72, | |
| "eval_steps_per_second": 3.196, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.4067167341709137, | |
| "eval_runtime": 0.6138, | |
| "eval_samples_per_second": 70.056, | |
| "eval_steps_per_second": 3.258, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.2461884319782257, | |
| "eval_runtime": 0.6216, | |
| "eval_samples_per_second": 69.175, | |
| "eval_steps_per_second": 3.217, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.2753083407878876, | |
| "eval_runtime": 0.6142, | |
| "eval_samples_per_second": 70.008, | |
| "eval_steps_per_second": 3.256, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.36990082263946533, | |
| "eval_runtime": 0.621, | |
| "eval_samples_per_second": 69.238, | |
| "eval_steps_per_second": 3.22, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 3.3077852725982666, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.1426, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.4983164668083191, | |
| "eval_runtime": 0.6111, | |
| "eval_samples_per_second": 70.367, | |
| "eval_steps_per_second": 3.273, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.31397297978401184, | |
| "eval_runtime": 0.6142, | |
| "eval_samples_per_second": 70.007, | |
| "eval_steps_per_second": 3.256, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.3470011055469513, | |
| "eval_runtime": 0.6168, | |
| "eval_samples_per_second": 69.71, | |
| "eval_steps_per_second": 3.242, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.4443318545818329, | |
| "eval_runtime": 0.6329, | |
| "eval_samples_per_second": 67.943, | |
| "eval_steps_per_second": 3.16, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.2582797110080719, | |
| "eval_runtime": 0.6186, | |
| "eval_samples_per_second": 69.513, | |
| "eval_steps_per_second": 3.233, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "grad_norm": 2.98862886428833, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.1385, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.22389179468154907, | |
| "eval_runtime": 0.617, | |
| "eval_samples_per_second": 69.689, | |
| "eval_steps_per_second": 3.241, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.2708161473274231, | |
| "eval_runtime": 0.6256, | |
| "eval_samples_per_second": 68.732, | |
| "eval_steps_per_second": 3.197, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.2660166621208191, | |
| "eval_runtime": 0.6333, | |
| "eval_samples_per_second": 67.897, | |
| "eval_steps_per_second": 3.158, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.27542343735694885, | |
| "eval_runtime": 0.6274, | |
| "eval_samples_per_second": 68.534, | |
| "eval_steps_per_second": 3.188, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.42455607652664185, | |
| "eval_runtime": 0.6136, | |
| "eval_samples_per_second": 70.083, | |
| "eval_steps_per_second": 3.26, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 2.3073959350585938, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.1202, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.2779475152492523, | |
| "eval_runtime": 0.6243, | |
| "eval_samples_per_second": 68.879, | |
| "eval_steps_per_second": 3.204, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.2725696265697479, | |
| "eval_runtime": 0.6116, | |
| "eval_samples_per_second": 70.313, | |
| "eval_steps_per_second": 3.27, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.25360527634620667, | |
| "eval_runtime": 0.6315, | |
| "eval_samples_per_second": 68.091, | |
| "eval_steps_per_second": 3.167, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.26673558354377747, | |
| "eval_runtime": 0.6196, | |
| "eval_samples_per_second": 69.396, | |
| "eval_steps_per_second": 3.228, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.4190940856933594, | |
| "eval_runtime": 0.6225, | |
| "eval_samples_per_second": 69.073, | |
| "eval_steps_per_second": 3.213, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "grad_norm": 5.0065016746521, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.1211, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.321259468793869, | |
| "eval_runtime": 0.6191, | |
| "eval_samples_per_second": 69.459, | |
| "eval_steps_per_second": 3.231, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.22903160750865936, | |
| "eval_runtime": 0.6149, | |
| "eval_samples_per_second": 69.933, | |
| "eval_steps_per_second": 3.253, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.30425024032592773, | |
| "eval_runtime": 0.6154, | |
| "eval_samples_per_second": 69.873, | |
| "eval_steps_per_second": 3.25, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.1906428039073944, | |
| "eval_runtime": 0.6363, | |
| "eval_samples_per_second": 67.575, | |
| "eval_steps_per_second": 3.143, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.32010775804519653, | |
| "eval_runtime": 0.6143, | |
| "eval_samples_per_second": 70.003, | |
| "eval_steps_per_second": 3.256, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 2.0770199298858643, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.1067, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.30616888403892517, | |
| "eval_runtime": 0.6546, | |
| "eval_samples_per_second": 65.687, | |
| "eval_steps_per_second": 3.055, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.2046656310558319, | |
| "eval_runtime": 0.6231, | |
| "eval_samples_per_second": 69.01, | |
| "eval_steps_per_second": 3.21, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.21158753335475922, | |
| "eval_runtime": 0.6461, | |
| "eval_samples_per_second": 66.557, | |
| "eval_steps_per_second": 3.096, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.2112913578748703, | |
| "eval_runtime": 0.6183, | |
| "eval_samples_per_second": 69.547, | |
| "eval_steps_per_second": 3.235, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.2339758723974228, | |
| "eval_runtime": 0.6307, | |
| "eval_samples_per_second": 68.174, | |
| "eval_steps_per_second": 3.171, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "grad_norm": 2.496302843093872, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0826, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.23282372951507568, | |
| "eval_runtime": 0.6179, | |
| "eval_samples_per_second": 69.589, | |
| "eval_steps_per_second": 3.237, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.2824217677116394, | |
| "eval_runtime": 0.6141, | |
| "eval_samples_per_second": 70.017, | |
| "eval_steps_per_second": 3.257, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.2921365201473236, | |
| "eval_runtime": 0.6146, | |
| "eval_samples_per_second": 69.966, | |
| "eval_steps_per_second": 3.254, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.26075631380081177, | |
| "eval_runtime": 0.6144, | |
| "eval_samples_per_second": 69.989, | |
| "eval_steps_per_second": 3.255, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.26498472690582275, | |
| "eval_runtime": 0.6164, | |
| "eval_samples_per_second": 69.761, | |
| "eval_steps_per_second": 3.245, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "grad_norm": 3.4850988388061523, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.0894, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.2878236472606659, | |
| "eval_runtime": 0.6314, | |
| "eval_samples_per_second": 68.108, | |
| "eval_steps_per_second": 3.168, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.29352495074272156, | |
| "eval_runtime": 0.6195, | |
| "eval_samples_per_second": 69.407, | |
| "eval_steps_per_second": 3.228, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.2655564546585083, | |
| "eval_runtime": 0.6137, | |
| "eval_samples_per_second": 70.063, | |
| "eval_steps_per_second": 3.259, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.3437863290309906, | |
| "eval_runtime": 0.6099, | |
| "eval_samples_per_second": 70.503, | |
| "eval_steps_per_second": 3.279, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.28401049971580505, | |
| "eval_runtime": 0.617, | |
| "eval_samples_per_second": 69.696, | |
| "eval_steps_per_second": 3.242, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "grad_norm": 2.939192295074463, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.0964, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.27114787697792053, | |
| "eval_runtime": 0.6121, | |
| "eval_samples_per_second": 70.255, | |
| "eval_steps_per_second": 3.268, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.2887828052043915, | |
| "eval_runtime": 0.6169, | |
| "eval_samples_per_second": 69.703, | |
| "eval_steps_per_second": 3.242, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.2722557783126831, | |
| "eval_runtime": 0.619, | |
| "eval_samples_per_second": 69.465, | |
| "eval_steps_per_second": 3.231, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.2562577426433563, | |
| "eval_runtime": 0.6161, | |
| "eval_samples_per_second": 69.795, | |
| "eval_steps_per_second": 3.246, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.2336176484823227, | |
| "eval_runtime": 0.6141, | |
| "eval_samples_per_second": 70.024, | |
| "eval_steps_per_second": 3.257, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "grad_norm": 2.394557476043701, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.0711, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.23857815563678741, | |
| "eval_runtime": 0.6236, | |
| "eval_samples_per_second": 68.958, | |
| "eval_steps_per_second": 3.207, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 71.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.24822531640529633, | |
| "eval_runtime": 0.612, | |
| "eval_samples_per_second": 70.257, | |
| "eval_steps_per_second": 3.268, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.28214532136917114, | |
| "eval_runtime": 0.6249, | |
| "eval_samples_per_second": 68.809, | |
| "eval_steps_per_second": 3.2, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.29410091042518616, | |
| "eval_runtime": 0.6147, | |
| "eval_samples_per_second": 69.95, | |
| "eval_steps_per_second": 3.253, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.2564258873462677, | |
| "eval_runtime": 0.623, | |
| "eval_samples_per_second": 69.022, | |
| "eval_steps_per_second": 3.21, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "grad_norm": 1.6818071603775024, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.0824, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.2509303092956543, | |
| "eval_runtime": 0.61, | |
| "eval_samples_per_second": 70.494, | |
| "eval_steps_per_second": 3.279, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.25438106060028076, | |
| "eval_runtime": 0.6085, | |
| "eval_samples_per_second": 70.661, | |
| "eval_steps_per_second": 3.287, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 77.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.24744297564029694, | |
| "eval_runtime": 0.6152, | |
| "eval_samples_per_second": 69.9, | |
| "eval_steps_per_second": 3.251, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.23753303289413452, | |
| "eval_runtime": 0.6291, | |
| "eval_samples_per_second": 68.355, | |
| "eval_steps_per_second": 3.179, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 79.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.23886720836162567, | |
| "eval_runtime": 0.627, | |
| "eval_samples_per_second": 68.58, | |
| "eval_steps_per_second": 3.19, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 1.1275471448898315, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.0691, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.23707760870456696, | |
| "eval_runtime": 0.6221, | |
| "eval_samples_per_second": 69.124, | |
| "eval_steps_per_second": 3.215, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 81.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.23934301733970642, | |
| "eval_runtime": 0.6174, | |
| "eval_samples_per_second": 69.646, | |
| "eval_steps_per_second": 3.239, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.25227802991867065, | |
| "eval_runtime": 0.6409, | |
| "eval_samples_per_second": 67.097, | |
| "eval_steps_per_second": 3.121, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 83.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.2676641345024109, | |
| "eval_runtime": 0.6354, | |
| "eval_samples_per_second": 67.676, | |
| "eval_steps_per_second": 3.148, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.29449349641799927, | |
| "eval_runtime": 0.645, | |
| "eval_samples_per_second": 66.667, | |
| "eval_steps_per_second": 3.101, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "grad_norm": 1.807352900505066, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.0638, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.3244798183441162, | |
| "eval_runtime": 0.6295, | |
| "eval_samples_per_second": 68.305, | |
| "eval_steps_per_second": 3.177, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.2960358262062073, | |
| "eval_runtime": 0.6257, | |
| "eval_samples_per_second": 68.718, | |
| "eval_steps_per_second": 3.196, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 87.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.2658250629901886, | |
| "eval_runtime": 0.6206, | |
| "eval_samples_per_second": 69.292, | |
| "eval_steps_per_second": 3.223, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.2614385485649109, | |
| "eval_runtime": 0.6167, | |
| "eval_samples_per_second": 69.721, | |
| "eval_steps_per_second": 3.243, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 89.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.26126545667648315, | |
| "eval_runtime": 0.6098, | |
| "eval_samples_per_second": 70.517, | |
| "eval_steps_per_second": 3.28, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "grad_norm": 2.1356263160705566, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0705, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.25489553809165955, | |
| "eval_runtime": 0.6147, | |
| "eval_samples_per_second": 69.953, | |
| "eval_steps_per_second": 3.254, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 91.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.2510479688644409, | |
| "eval_runtime": 0.6176, | |
| "eval_samples_per_second": 69.628, | |
| "eval_steps_per_second": 3.238, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.25144004821777344, | |
| "eval_runtime": 0.6229, | |
| "eval_samples_per_second": 69.032, | |
| "eval_steps_per_second": 3.211, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 93.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.25218266248703003, | |
| "eval_runtime": 0.6238, | |
| "eval_samples_per_second": 68.933, | |
| "eval_steps_per_second": 3.206, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 94.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.2504193186759949, | |
| "eval_runtime": 0.6666, | |
| "eval_samples_per_second": 64.51, | |
| "eval_steps_per_second": 3.0, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "grad_norm": 1.0901120901107788, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.0666, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.2458992600440979, | |
| "eval_runtime": 0.6193, | |
| "eval_samples_per_second": 69.433, | |
| "eval_steps_per_second": 3.229, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.24242638051509857, | |
| "eval_runtime": 0.6296, | |
| "eval_samples_per_second": 68.301, | |
| "eval_steps_per_second": 3.177, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 97.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.2427695095539093, | |
| "eval_runtime": 0.6141, | |
| "eval_samples_per_second": 70.016, | |
| "eval_steps_per_second": 3.257, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 98.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.24509721994400024, | |
| "eval_runtime": 0.6364, | |
| "eval_samples_per_second": 67.568, | |
| "eval_steps_per_second": 3.143, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.24713072180747986, | |
| "eval_runtime": 0.6105, | |
| "eval_samples_per_second": 70.437, | |
| "eval_steps_per_second": 3.276, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "grad_norm": 2.5352766513824463, | |
| "learning_rate": 0.0, | |
| "loss": 0.0566, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.24816666543483734, | |
| "eval_runtime": 0.6223, | |
| "eval_samples_per_second": 69.104, | |
| "eval_steps_per_second": 3.214, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "step": 200, | |
| "total_flos": 1.867590382823424e+18, | |
| "train_loss": 0.15699556827545166, | |
| "train_runtime": 1120.5665, | |
| "train_samples_per_second": 21.507, | |
| "train_steps_per_second": 0.178 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.26246392726898193, | |
| "eval_runtime": 0.6852, | |
| "eval_samples_per_second": 62.754, | |
| "eval_steps_per_second": 2.919, | |
| "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 | |
| } | |