| { | |
| "best_metric": 0.9328358208955224, | |
| "best_model_checkpoint": "/kaggle/working/swin-brain-tumor-type-classification/checkpoint-627", | |
| "epoch": 35.0, | |
| "eval_steps": 500, | |
| "global_step": 665, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5263157894736842, | |
| "grad_norm": 6.779318809509277, | |
| "learning_rate": 7.4626865671641785e-06, | |
| "loss": 2.6739, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.19776119402985073, | |
| "eval_loss": 2.423863410949707, | |
| "eval_runtime": 2.1298, | |
| "eval_samples_per_second": 125.831, | |
| "eval_steps_per_second": 4.226, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.0526315789473684, | |
| "grad_norm": 5.687614917755127, | |
| "learning_rate": 1.4925373134328357e-05, | |
| "loss": 2.4803, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.5789473684210527, | |
| "grad_norm": 7.469175338745117, | |
| "learning_rate": 2.238805970149254e-05, | |
| "loss": 2.2793, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.3619402985074627, | |
| "eval_loss": 2.101121187210083, | |
| "eval_runtime": 2.1211, | |
| "eval_samples_per_second": 126.35, | |
| "eval_steps_per_second": 4.243, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 2.1052631578947367, | |
| "grad_norm": 7.969486713409424, | |
| "learning_rate": 2.9850746268656714e-05, | |
| "loss": 2.1009, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.6315789473684212, | |
| "grad_norm": 11.056779861450195, | |
| "learning_rate": 3.73134328358209e-05, | |
| "loss": 1.9273, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.4962686567164179, | |
| "eval_loss": 1.6876119375228882, | |
| "eval_runtime": 2.1801, | |
| "eval_samples_per_second": 122.928, | |
| "eval_steps_per_second": 4.128, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 3.1578947368421053, | |
| "grad_norm": 7.5373334884643555, | |
| "learning_rate": 4.477611940298508e-05, | |
| "loss": 1.6252, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.6842105263157894, | |
| "grad_norm": 10.007360458374023, | |
| "learning_rate": 4.974916387959867e-05, | |
| "loss": 1.5549, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5708955223880597, | |
| "eval_loss": 1.3625893592834473, | |
| "eval_runtime": 2.2749, | |
| "eval_samples_per_second": 117.809, | |
| "eval_steps_per_second": 3.956, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 4.2105263157894735, | |
| "grad_norm": 10.789587020874023, | |
| "learning_rate": 4.891304347826087e-05, | |
| "loss": 1.3971, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.7368421052631575, | |
| "grad_norm": 9.604104042053223, | |
| "learning_rate": 4.8076923076923084e-05, | |
| "loss": 1.2386, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6268656716417911, | |
| "eval_loss": 1.1268030405044556, | |
| "eval_runtime": 2.5243, | |
| "eval_samples_per_second": 106.167, | |
| "eval_steps_per_second": 3.565, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 5.2631578947368425, | |
| "grad_norm": 12.851019859313965, | |
| "learning_rate": 4.724080267558529e-05, | |
| "loss": 1.1166, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.7894736842105265, | |
| "grad_norm": 11.019944190979004, | |
| "learning_rate": 4.640468227424749e-05, | |
| "loss": 1.1004, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7014925373134329, | |
| "eval_loss": 0.9444876909255981, | |
| "eval_runtime": 2.1665, | |
| "eval_samples_per_second": 123.7, | |
| "eval_steps_per_second": 4.154, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 6.315789473684211, | |
| "grad_norm": 14.164745330810547, | |
| "learning_rate": 4.55685618729097e-05, | |
| "loss": 1.0225, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.842105263157895, | |
| "grad_norm": 9.702640533447266, | |
| "learning_rate": 4.473244147157191e-05, | |
| "loss": 0.9008, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7313432835820896, | |
| "eval_loss": 0.8700507879257202, | |
| "eval_runtime": 2.1524, | |
| "eval_samples_per_second": 124.514, | |
| "eval_steps_per_second": 4.181, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 7.368421052631579, | |
| "grad_norm": 12.368742942810059, | |
| "learning_rate": 4.3896321070234115e-05, | |
| "loss": 0.8022, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 7.894736842105263, | |
| "grad_norm": 16.489498138427734, | |
| "learning_rate": 4.306020066889632e-05, | |
| "loss": 0.8023, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7425373134328358, | |
| "eval_loss": 0.7917194962501526, | |
| "eval_runtime": 2.1698, | |
| "eval_samples_per_second": 123.513, | |
| "eval_steps_per_second": 4.148, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 8.421052631578947, | |
| "grad_norm": 11.644203186035156, | |
| "learning_rate": 4.222408026755853e-05, | |
| "loss": 0.8047, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 8.947368421052632, | |
| "grad_norm": 10.630133628845215, | |
| "learning_rate": 4.1387959866220736e-05, | |
| "loss": 0.6566, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.7611940298507462, | |
| "eval_loss": 0.6989582777023315, | |
| "eval_runtime": 2.1317, | |
| "eval_samples_per_second": 125.721, | |
| "eval_steps_per_second": 4.222, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 9.473684210526315, | |
| "grad_norm": 9.887234687805176, | |
| "learning_rate": 4.055183946488295e-05, | |
| "loss": 0.6759, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 16.37430191040039, | |
| "learning_rate": 3.971571906354515e-05, | |
| "loss": 0.6691, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7985074626865671, | |
| "eval_loss": 0.6203530430793762, | |
| "eval_runtime": 2.1249, | |
| "eval_samples_per_second": 126.122, | |
| "eval_steps_per_second": 4.235, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 10.526315789473685, | |
| "grad_norm": 12.469463348388672, | |
| "learning_rate": 3.8879598662207364e-05, | |
| "loss": 0.5605, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.832089552238806, | |
| "eval_loss": 0.5511359572410583, | |
| "eval_runtime": 2.1778, | |
| "eval_samples_per_second": 123.061, | |
| "eval_steps_per_second": 4.133, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 11.052631578947368, | |
| "grad_norm": 12.518712043762207, | |
| "learning_rate": 3.804347826086957e-05, | |
| "loss": 0.5599, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 11.578947368421053, | |
| "grad_norm": 9.64068603515625, | |
| "learning_rate": 3.7207357859531773e-05, | |
| "loss": 0.5472, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.835820895522388, | |
| "eval_loss": 0.49448007345199585, | |
| "eval_runtime": 2.1908, | |
| "eval_samples_per_second": 122.327, | |
| "eval_steps_per_second": 4.108, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 12.105263157894736, | |
| "grad_norm": 8.733270645141602, | |
| "learning_rate": 3.637123745819398e-05, | |
| "loss": 0.4963, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 12.631578947368421, | |
| "grad_norm": 13.732227325439453, | |
| "learning_rate": 3.553511705685619e-05, | |
| "loss": 0.5098, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8619402985074627, | |
| "eval_loss": 0.4302270710468292, | |
| "eval_runtime": 2.1349, | |
| "eval_samples_per_second": 125.533, | |
| "eval_steps_per_second": 4.216, | |
| "step": 247 | |
| }, | |
| { | |
| "epoch": 13.157894736842104, | |
| "grad_norm": 8.828410148620605, | |
| "learning_rate": 3.4698996655518395e-05, | |
| "loss": 0.439, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 13.68421052631579, | |
| "grad_norm": 11.607008934020996, | |
| "learning_rate": 3.3862876254180606e-05, | |
| "loss": 0.4362, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8843283582089553, | |
| "eval_loss": 0.40266644954681396, | |
| "eval_runtime": 2.121, | |
| "eval_samples_per_second": 126.357, | |
| "eval_steps_per_second": 4.243, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 14.210526315789474, | |
| "grad_norm": 7.250407695770264, | |
| "learning_rate": 3.302675585284281e-05, | |
| "loss": 0.4385, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 14.736842105263158, | |
| "grad_norm": 9.448441505432129, | |
| "learning_rate": 3.2190635451505016e-05, | |
| "loss": 0.416, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8656716417910447, | |
| "eval_loss": 0.39556798338890076, | |
| "eval_runtime": 2.1456, | |
| "eval_samples_per_second": 124.908, | |
| "eval_steps_per_second": 4.195, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 15.263157894736842, | |
| "grad_norm": 7.7804274559021, | |
| "learning_rate": 3.135451505016723e-05, | |
| "loss": 0.3697, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 15.789473684210526, | |
| "grad_norm": 14.305925369262695, | |
| "learning_rate": 3.051839464882943e-05, | |
| "loss": 0.4095, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8880597014925373, | |
| "eval_loss": 0.3604812026023865, | |
| "eval_runtime": 2.1246, | |
| "eval_samples_per_second": 126.144, | |
| "eval_steps_per_second": 4.236, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 16.31578947368421, | |
| "grad_norm": 11.360480308532715, | |
| "learning_rate": 2.9682274247491644e-05, | |
| "loss": 0.3991, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 16.842105263157894, | |
| "grad_norm": 11.096769332885742, | |
| "learning_rate": 2.8846153846153845e-05, | |
| "loss": 0.3577, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.8917910447761194, | |
| "eval_loss": 0.3339410424232483, | |
| "eval_runtime": 2.1207, | |
| "eval_samples_per_second": 126.373, | |
| "eval_steps_per_second": 4.244, | |
| "step": 323 | |
| }, | |
| { | |
| "epoch": 17.36842105263158, | |
| "grad_norm": 9.879475593566895, | |
| "learning_rate": 2.8010033444816054e-05, | |
| "loss": 0.291, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 17.894736842105264, | |
| "grad_norm": 15.603042602539062, | |
| "learning_rate": 2.7173913043478262e-05, | |
| "loss": 0.3624, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8694029850746269, | |
| "eval_loss": 0.38826432824134827, | |
| "eval_runtime": 2.2245, | |
| "eval_samples_per_second": 120.479, | |
| "eval_steps_per_second": 4.046, | |
| "step": 342 | |
| }, | |
| { | |
| "epoch": 18.42105263157895, | |
| "grad_norm": 12.606857299804688, | |
| "learning_rate": 2.633779264214047e-05, | |
| "loss": 0.3399, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 18.94736842105263, | |
| "grad_norm": 12.286445617675781, | |
| "learning_rate": 2.5501672240802675e-05, | |
| "loss": 0.304, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.8768656716417911, | |
| "eval_loss": 0.3496492803096771, | |
| "eval_runtime": 2.1504, | |
| "eval_samples_per_second": 124.629, | |
| "eval_steps_per_second": 4.185, | |
| "step": 361 | |
| }, | |
| { | |
| "epoch": 19.473684210526315, | |
| "grad_norm": 8.954345703125, | |
| "learning_rate": 2.4665551839464883e-05, | |
| "loss": 0.3156, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 30.505029678344727, | |
| "learning_rate": 2.382943143812709e-05, | |
| "loss": 0.2784, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8805970149253731, | |
| "eval_loss": 0.32751259207725525, | |
| "eval_runtime": 2.1365, | |
| "eval_samples_per_second": 125.441, | |
| "eval_steps_per_second": 4.213, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 20.526315789473685, | |
| "grad_norm": 6.59121036529541, | |
| "learning_rate": 2.29933110367893e-05, | |
| "loss": 0.2763, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.8805970149253731, | |
| "eval_loss": 0.3721315860748291, | |
| "eval_runtime": 2.1701, | |
| "eval_samples_per_second": 123.499, | |
| "eval_steps_per_second": 4.147, | |
| "step": 399 | |
| }, | |
| { | |
| "epoch": 21.05263157894737, | |
| "grad_norm": 6.725207328796387, | |
| "learning_rate": 2.2157190635451504e-05, | |
| "loss": 0.2903, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 21.57894736842105, | |
| "grad_norm": 12.423023223876953, | |
| "learning_rate": 2.1321070234113713e-05, | |
| "loss": 0.2824, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.8955223880597015, | |
| "eval_loss": 0.31556689739227295, | |
| "eval_runtime": 2.1623, | |
| "eval_samples_per_second": 123.942, | |
| "eval_steps_per_second": 4.162, | |
| "step": 418 | |
| }, | |
| { | |
| "epoch": 22.105263157894736, | |
| "grad_norm": 7.236621856689453, | |
| "learning_rate": 2.048494983277592e-05, | |
| "loss": 0.2299, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 22.63157894736842, | |
| "grad_norm": 6.276068210601807, | |
| "learning_rate": 1.964882943143813e-05, | |
| "loss": 0.2453, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.8843283582089553, | |
| "eval_loss": 0.3155025541782379, | |
| "eval_runtime": 2.1435, | |
| "eval_samples_per_second": 125.029, | |
| "eval_steps_per_second": 4.199, | |
| "step": 437 | |
| }, | |
| { | |
| "epoch": 23.157894736842106, | |
| "grad_norm": 6.001166820526123, | |
| "learning_rate": 1.8812709030100337e-05, | |
| "loss": 0.248, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 23.68421052631579, | |
| "grad_norm": 9.594138145446777, | |
| "learning_rate": 1.7976588628762542e-05, | |
| "loss": 0.2438, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.9029850746268657, | |
| "eval_loss": 0.2927631437778473, | |
| "eval_runtime": 2.2056, | |
| "eval_samples_per_second": 121.509, | |
| "eval_steps_per_second": 4.081, | |
| "step": 456 | |
| }, | |
| { | |
| "epoch": 24.210526315789473, | |
| "grad_norm": 12.752880096435547, | |
| "learning_rate": 1.714046822742475e-05, | |
| "loss": 0.2472, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 24.736842105263158, | |
| "grad_norm": 5.313000202178955, | |
| "learning_rate": 1.630434782608696e-05, | |
| "loss": 0.2285, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.9216417910447762, | |
| "eval_loss": 0.26671093702316284, | |
| "eval_runtime": 2.2227, | |
| "eval_samples_per_second": 120.573, | |
| "eval_steps_per_second": 4.049, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 25.263157894736842, | |
| "grad_norm": 8.309535026550293, | |
| "learning_rate": 1.5468227424749167e-05, | |
| "loss": 0.2555, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 25.789473684210527, | |
| "grad_norm": 7.868557453155518, | |
| "learning_rate": 1.4632107023411373e-05, | |
| "loss": 0.2478, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.914179104477612, | |
| "eval_loss": 0.2815816402435303, | |
| "eval_runtime": 2.1383, | |
| "eval_samples_per_second": 125.336, | |
| "eval_steps_per_second": 4.209, | |
| "step": 494 | |
| }, | |
| { | |
| "epoch": 26.31578947368421, | |
| "grad_norm": 6.559621334075928, | |
| "learning_rate": 1.3795986622073578e-05, | |
| "loss": 0.2601, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 26.842105263157894, | |
| "grad_norm": 7.318304538726807, | |
| "learning_rate": 1.2959866220735786e-05, | |
| "loss": 0.2242, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.8992537313432836, | |
| "eval_loss": 0.2767724096775055, | |
| "eval_runtime": 2.1721, | |
| "eval_samples_per_second": 123.385, | |
| "eval_steps_per_second": 4.144, | |
| "step": 513 | |
| }, | |
| { | |
| "epoch": 27.36842105263158, | |
| "grad_norm": 9.232494354248047, | |
| "learning_rate": 1.2123745819397994e-05, | |
| "loss": 0.1988, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 27.894736842105264, | |
| "grad_norm": 6.716543197631836, | |
| "learning_rate": 1.1287625418060201e-05, | |
| "loss": 0.2, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.914179104477612, | |
| "eval_loss": 0.2814846932888031, | |
| "eval_runtime": 2.2098, | |
| "eval_samples_per_second": 121.279, | |
| "eval_steps_per_second": 4.073, | |
| "step": 532 | |
| }, | |
| { | |
| "epoch": 28.42105263157895, | |
| "grad_norm": 9.893804550170898, | |
| "learning_rate": 1.045150501672241e-05, | |
| "loss": 0.2091, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 28.94736842105263, | |
| "grad_norm": 9.348865509033203, | |
| "learning_rate": 9.615384615384616e-06, | |
| "loss": 0.2076, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.9216417910447762, | |
| "eval_loss": 0.24431268870830536, | |
| "eval_runtime": 2.148, | |
| "eval_samples_per_second": 124.765, | |
| "eval_steps_per_second": 4.19, | |
| "step": 551 | |
| }, | |
| { | |
| "epoch": 29.473684210526315, | |
| "grad_norm": 8.77604866027832, | |
| "learning_rate": 8.779264214046822e-06, | |
| "loss": 0.2026, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 14.116354942321777, | |
| "learning_rate": 7.94314381270903e-06, | |
| "loss": 0.1978, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.9216417910447762, | |
| "eval_loss": 0.23809637129306793, | |
| "eval_runtime": 2.1703, | |
| "eval_samples_per_second": 123.486, | |
| "eval_steps_per_second": 4.147, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 30.526315789473685, | |
| "grad_norm": 11.3234281539917, | |
| "learning_rate": 7.107023411371237e-06, | |
| "loss": 0.1821, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy": 0.9216417910447762, | |
| "eval_loss": 0.25633105635643005, | |
| "eval_runtime": 2.1804, | |
| "eval_samples_per_second": 122.916, | |
| "eval_steps_per_second": 4.128, | |
| "step": 589 | |
| }, | |
| { | |
| "epoch": 31.05263157894737, | |
| "grad_norm": 7.66957950592041, | |
| "learning_rate": 6.270903010033445e-06, | |
| "loss": 0.2099, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 31.57894736842105, | |
| "grad_norm": 5.773667335510254, | |
| "learning_rate": 5.4347826086956525e-06, | |
| "loss": 0.1786, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.9253731343283582, | |
| "eval_loss": 0.24489234387874603, | |
| "eval_runtime": 2.2643, | |
| "eval_samples_per_second": 118.359, | |
| "eval_steps_per_second": 3.975, | |
| "step": 608 | |
| }, | |
| { | |
| "epoch": 32.10526315789474, | |
| "grad_norm": 5.752429008483887, | |
| "learning_rate": 4.598662207357859e-06, | |
| "loss": 0.1785, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 32.63157894736842, | |
| "grad_norm": 9.643966674804688, | |
| "learning_rate": 3.7625418060200673e-06, | |
| "loss": 0.1809, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy": 0.9328358208955224, | |
| "eval_loss": 0.23848846554756165, | |
| "eval_runtime": 2.3814, | |
| "eval_samples_per_second": 112.539, | |
| "eval_steps_per_second": 3.779, | |
| "step": 627 | |
| }, | |
| { | |
| "epoch": 33.1578947368421, | |
| "grad_norm": 8.57278823852539, | |
| "learning_rate": 2.9264214046822746e-06, | |
| "loss": 0.1825, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 33.68421052631579, | |
| "grad_norm": 6.0924296379089355, | |
| "learning_rate": 2.0903010033444816e-06, | |
| "loss": 0.1812, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.9291044776119403, | |
| "eval_loss": 0.2448083907365799, | |
| "eval_runtime": 2.3113, | |
| "eval_samples_per_second": 115.954, | |
| "eval_steps_per_second": 3.894, | |
| "step": 646 | |
| }, | |
| { | |
| "epoch": 34.21052631578947, | |
| "grad_norm": 11.392114639282227, | |
| "learning_rate": 1.254180602006689e-06, | |
| "loss": 0.2006, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 34.73684210526316, | |
| "grad_norm": 7.01861047744751, | |
| "learning_rate": 4.180602006688963e-07, | |
| "loss": 0.1688, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy": 0.9291044776119403, | |
| "eval_loss": 0.24462540447711945, | |
| "eval_runtime": 2.2089, | |
| "eval_samples_per_second": 121.33, | |
| "eval_steps_per_second": 4.074, | |
| "step": 665 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "step": 665, | |
| "total_flos": 2.0920841725722624e+18, | |
| "train_loss": 0.6001271831361871, | |
| "train_runtime": 1153.5131, | |
| "train_samples_per_second": 72.942, | |
| "train_steps_per_second": 0.576 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 665, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 35, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "CustomEarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 1, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.0920841725722624e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |