| { | |
| "best_metric": 0.8548387096774194, | |
| "best_model_checkpoint": "swiftformer-xs-OT\\checkpoint-139", | |
| "epoch": 35.55555555555556, | |
| "eval_steps": 500, | |
| "global_step": 160, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.89, | |
| "eval_accuracy": 0.5161290322580645, | |
| "eval_loss": 1.3804326057434082, | |
| "eval_runtime": 2.0395, | |
| "eval_samples_per_second": 30.4, | |
| "eval_steps_per_second": 1.961, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.532258064516129, | |
| "eval_loss": 1.2554447650909424, | |
| "eval_runtime": 1.9685, | |
| "eval_samples_per_second": 31.497, | |
| "eval_steps_per_second": 2.032, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 0.0009375, | |
| "loss": 1.3469, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "eval_accuracy": 0.6612903225806451, | |
| "eval_loss": 0.9724773168563843, | |
| "eval_runtime": 1.9795, | |
| "eval_samples_per_second": 31.321, | |
| "eval_steps_per_second": 2.021, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7580645161290323, | |
| "eval_loss": 0.7085855007171631, | |
| "eval_runtime": 2.3836, | |
| "eval_samples_per_second": 26.011, | |
| "eval_steps_per_second": 1.678, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 0.0014583333333333334, | |
| "loss": 0.9831, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "eval_accuracy": 0.7258064516129032, | |
| "eval_loss": 0.8855839371681213, | |
| "eval_runtime": 2.156, | |
| "eval_samples_per_second": 28.757, | |
| "eval_steps_per_second": 1.855, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7580645161290323, | |
| "eval_loss": 0.7723659873008728, | |
| "eval_runtime": 1.9627, | |
| "eval_samples_per_second": 31.588, | |
| "eval_steps_per_second": 2.038, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 0.0013541666666666667, | |
| "loss": 0.7441, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 6.89, | |
| "eval_accuracy": 0.7258064516129032, | |
| "eval_loss": 0.819038450717926, | |
| "eval_runtime": 1.979, | |
| "eval_samples_per_second": 31.33, | |
| "eval_steps_per_second": 2.021, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7741935483870968, | |
| "eval_loss": 0.6897338628768921, | |
| "eval_runtime": 1.8919, | |
| "eval_samples_per_second": 32.77, | |
| "eval_steps_per_second": 2.114, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "learning_rate": 0.00125, | |
| "loss": 0.6939, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "eval_accuracy": 0.7258064516129032, | |
| "eval_loss": 0.6599262356758118, | |
| "eval_runtime": 1.9435, | |
| "eval_samples_per_second": 31.902, | |
| "eval_steps_per_second": 2.058, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7741935483870968, | |
| "eval_loss": 0.6288474798202515, | |
| "eval_runtime": 2.1175, | |
| "eval_samples_per_second": 29.28, | |
| "eval_steps_per_second": 1.889, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 10.89, | |
| "eval_accuracy": 0.7580645161290323, | |
| "eval_loss": 0.6333299279212952, | |
| "eval_runtime": 2.055, | |
| "eval_samples_per_second": 30.17, | |
| "eval_steps_per_second": 1.946, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 11.11, | |
| "learning_rate": 0.0011458333333333333, | |
| "loss": 0.5861, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7741935483870968, | |
| "eval_loss": 0.6206099987030029, | |
| "eval_runtime": 2.139, | |
| "eval_samples_per_second": 28.985, | |
| "eval_steps_per_second": 1.87, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 12.89, | |
| "eval_accuracy": 0.7903225806451613, | |
| "eval_loss": 0.5262896418571472, | |
| "eval_runtime": 2.0025, | |
| "eval_samples_per_second": 30.961, | |
| "eval_steps_per_second": 1.998, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 13.33, | |
| "learning_rate": 0.0010416666666666667, | |
| "loss": 0.5018, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8064516129032258, | |
| "eval_loss": 0.583601176738739, | |
| "eval_runtime": 2.0815, | |
| "eval_samples_per_second": 29.786, | |
| "eval_steps_per_second": 1.922, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 14.89, | |
| "eval_accuracy": 0.7419354838709677, | |
| "eval_loss": 0.6125115156173706, | |
| "eval_runtime": 1.948, | |
| "eval_samples_per_second": 31.828, | |
| "eval_steps_per_second": 2.053, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 15.56, | |
| "learning_rate": 0.0009375, | |
| "loss": 0.4642, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8064516129032258, | |
| "eval_loss": 0.5431403517723083, | |
| "eval_runtime": 1.921, | |
| "eval_samples_per_second": 32.276, | |
| "eval_steps_per_second": 2.082, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 16.89, | |
| "eval_accuracy": 0.8387096774193549, | |
| "eval_loss": 0.5893498063087463, | |
| "eval_runtime": 1.9645, | |
| "eval_samples_per_second": 31.561, | |
| "eval_steps_per_second": 2.036, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 17.78, | |
| "learning_rate": 0.0008333333333333334, | |
| "loss": 0.4064, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8064516129032258, | |
| "eval_loss": 0.49968525767326355, | |
| "eval_runtime": 1.8915, | |
| "eval_samples_per_second": 32.779, | |
| "eval_steps_per_second": 2.115, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 18.89, | |
| "eval_accuracy": 0.7741935483870968, | |
| "eval_loss": 0.5474066138267517, | |
| "eval_runtime": 1.968, | |
| "eval_samples_per_second": 31.505, | |
| "eval_steps_per_second": 2.033, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 0.0007291666666666667, | |
| "loss": 0.4275, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7903225806451613, | |
| "eval_loss": 0.6748386025428772, | |
| "eval_runtime": 1.963, | |
| "eval_samples_per_second": 31.585, | |
| "eval_steps_per_second": 2.038, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 20.89, | |
| "eval_accuracy": 0.7580645161290323, | |
| "eval_loss": 0.6369444131851196, | |
| "eval_runtime": 1.98, | |
| "eval_samples_per_second": 31.314, | |
| "eval_steps_per_second": 2.02, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.7741935483870968, | |
| "eval_loss": 0.5609545111656189, | |
| "eval_runtime": 1.974, | |
| "eval_samples_per_second": 31.409, | |
| "eval_steps_per_second": 2.026, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 22.22, | |
| "learning_rate": 0.000625, | |
| "loss": 0.373, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 22.89, | |
| "eval_accuracy": 0.7903225806451613, | |
| "eval_loss": 0.5259799957275391, | |
| "eval_runtime": 1.9815, | |
| "eval_samples_per_second": 31.29, | |
| "eval_steps_per_second": 2.019, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.8387096774193549, | |
| "eval_loss": 0.5416117310523987, | |
| "eval_runtime": 2.1095, | |
| "eval_samples_per_second": 29.391, | |
| "eval_steps_per_second": 1.896, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 24.44, | |
| "learning_rate": 0.0005208333333333333, | |
| "loss": 0.2931, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 24.89, | |
| "eval_accuracy": 0.8387096774193549, | |
| "eval_loss": 0.5146054029464722, | |
| "eval_runtime": 2.1155, | |
| "eval_samples_per_second": 29.308, | |
| "eval_steps_per_second": 1.891, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.7741935483870968, | |
| "eval_loss": 0.5180203914642334, | |
| "eval_runtime": 1.8704, | |
| "eval_samples_per_second": 33.147, | |
| "eval_steps_per_second": 2.139, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 26.67, | |
| "learning_rate": 0.0004166666666666667, | |
| "loss": 0.3135, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 26.89, | |
| "eval_accuracy": 0.8225806451612904, | |
| "eval_loss": 0.5169466733932495, | |
| "eval_runtime": 2.0975, | |
| "eval_samples_per_second": 29.559, | |
| "eval_steps_per_second": 1.907, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.8387096774193549, | |
| "eval_loss": 0.5491407513618469, | |
| "eval_runtime": 2.1065, | |
| "eval_samples_per_second": 29.433, | |
| "eval_steps_per_second": 1.899, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 28.89, | |
| "learning_rate": 0.0003125, | |
| "loss": 0.2342, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 28.89, | |
| "eval_accuracy": 0.8387096774193549, | |
| "eval_loss": 0.5384820699691772, | |
| "eval_runtime": 2.084, | |
| "eval_samples_per_second": 29.751, | |
| "eval_steps_per_second": 1.919, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.8387096774193549, | |
| "eval_loss": 0.5455949306488037, | |
| "eval_runtime": 1.9334, | |
| "eval_samples_per_second": 32.067, | |
| "eval_steps_per_second": 2.069, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 30.89, | |
| "eval_accuracy": 0.8548387096774194, | |
| "eval_loss": 0.4955712854862213, | |
| "eval_runtime": 2.2074, | |
| "eval_samples_per_second": 28.088, | |
| "eval_steps_per_second": 1.812, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 31.11, | |
| "learning_rate": 0.00020833333333333335, | |
| "loss": 0.2411, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.8225806451612904, | |
| "eval_loss": 0.5254048109054565, | |
| "eval_runtime": 2.3185, | |
| "eval_samples_per_second": 26.741, | |
| "eval_steps_per_second": 1.725, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 32.89, | |
| "eval_accuracy": 0.8387096774193549, | |
| "eval_loss": 0.5533136129379272, | |
| "eval_runtime": 1.8899, | |
| "eval_samples_per_second": 32.805, | |
| "eval_steps_per_second": 2.116, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "learning_rate": 0.00010416666666666667, | |
| "loss": 0.2135, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.8387096774193549, | |
| "eval_loss": 0.5612671971321106, | |
| "eval_runtime": 2.0095, | |
| "eval_samples_per_second": 30.854, | |
| "eval_steps_per_second": 1.991, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 34.89, | |
| "eval_accuracy": 0.8225806451612904, | |
| "eval_loss": 0.5748429894447327, | |
| "eval_runtime": 2.0385, | |
| "eval_samples_per_second": 30.415, | |
| "eval_steps_per_second": 1.962, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 35.56, | |
| "learning_rate": 0.0, | |
| "loss": 0.1904, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 35.56, | |
| "eval_accuracy": 0.8387096774193549, | |
| "eval_loss": 0.5843761563301086, | |
| "eval_runtime": 2.02, | |
| "eval_samples_per_second": 30.694, | |
| "eval_steps_per_second": 1.98, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 35.56, | |
| "step": 160, | |
| "total_flos": 2.807020017156096e+16, | |
| "train_loss": 0.5008050501346588, | |
| "train_runtime": 452.9272, | |
| "train_samples_per_second": 25.435, | |
| "train_steps_per_second": 0.353 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 160, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 40, | |
| "save_steps": 500, | |
| "total_flos": 2.807020017156096e+16, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |