|
{ |
|
"best_metric": 0.5454545454545454, |
|
"best_model_checkpoint": "swin-small-patch4-window7-224-finetuned-isic217/checkpoint-269", |
|
"epoch": 29.387755102040817, |
|
"eval_steps": 500, |
|
"global_step": 720, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.40816326530612246, |
|
"grad_norm": 15.3554048538208, |
|
"learning_rate": 6.944444444444445e-06, |
|
"loss": 2.1982, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.8163265306122449, |
|
"grad_norm": 12.736021041870117, |
|
"learning_rate": 1.388888888888889e-05, |
|
"loss": 2.1844, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.9795918367346939, |
|
"eval_accuracy": 0.13636363636363635, |
|
"eval_loss": 2.1102521419525146, |
|
"eval_runtime": 4.2332, |
|
"eval_samples_per_second": 5.197, |
|
"eval_steps_per_second": 2.599, |
|
"step": 24 |
|
}, |
|
{ |
|
"epoch": 1.2244897959183674, |
|
"grad_norm": 20.908735275268555, |
|
"learning_rate": 2.0833333333333336e-05, |
|
"loss": 2.0984, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 1.6326530612244898, |
|
"grad_norm": 17.49510955810547, |
|
"learning_rate": 2.777777777777778e-05, |
|
"loss": 2.0018, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.2727272727272727, |
|
"eval_loss": 1.8736993074417114, |
|
"eval_runtime": 4.0949, |
|
"eval_samples_per_second": 5.373, |
|
"eval_steps_per_second": 2.686, |
|
"step": 49 |
|
}, |
|
{ |
|
"epoch": 2.0408163265306123, |
|
"grad_norm": 16.44957733154297, |
|
"learning_rate": 3.472222222222222e-05, |
|
"loss": 1.9104, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 2.4489795918367347, |
|
"grad_norm": 25.78542137145996, |
|
"learning_rate": 4.166666666666667e-05, |
|
"loss": 1.6759, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 2.857142857142857, |
|
"grad_norm": 35.257904052734375, |
|
"learning_rate": 4.8611111111111115e-05, |
|
"loss": 1.6474, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 2.979591836734694, |
|
"eval_accuracy": 0.2727272727272727, |
|
"eval_loss": 1.901879906654358, |
|
"eval_runtime": 3.9033, |
|
"eval_samples_per_second": 5.636, |
|
"eval_steps_per_second": 2.818, |
|
"step": 73 |
|
}, |
|
{ |
|
"epoch": 3.2653061224489797, |
|
"grad_norm": 24.751829147338867, |
|
"learning_rate": 4.938271604938271e-05, |
|
"loss": 1.5587, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 3.673469387755102, |
|
"grad_norm": 29.13814353942871, |
|
"learning_rate": 4.8611111111111115e-05, |
|
"loss": 1.3757, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.36363636363636365, |
|
"eval_loss": 1.7486611604690552, |
|
"eval_runtime": 4.8083, |
|
"eval_samples_per_second": 4.575, |
|
"eval_steps_per_second": 2.288, |
|
"step": 98 |
|
}, |
|
{ |
|
"epoch": 4.081632653061225, |
|
"grad_norm": 24.104280471801758, |
|
"learning_rate": 4.783950617283951e-05, |
|
"loss": 1.3791, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 4.489795918367347, |
|
"grad_norm": 24.011398315429688, |
|
"learning_rate": 4.70679012345679e-05, |
|
"loss": 1.0546, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 4.8979591836734695, |
|
"grad_norm": 18.644390106201172, |
|
"learning_rate": 4.62962962962963e-05, |
|
"loss": 1.1526, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 4.979591836734694, |
|
"eval_accuracy": 0.4090909090909091, |
|
"eval_loss": 1.7576110363006592, |
|
"eval_runtime": 3.8964, |
|
"eval_samples_per_second": 5.646, |
|
"eval_steps_per_second": 2.823, |
|
"step": 122 |
|
}, |
|
{ |
|
"epoch": 5.3061224489795915, |
|
"grad_norm": 19.769208908081055, |
|
"learning_rate": 4.5524691358024696e-05, |
|
"loss": 0.7525, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 5.714285714285714, |
|
"grad_norm": 28.662109375, |
|
"learning_rate": 4.4753086419753084e-05, |
|
"loss": 0.9161, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_accuracy": 0.5, |
|
"eval_loss": 1.5886327028274536, |
|
"eval_runtime": 4.047, |
|
"eval_samples_per_second": 5.436, |
|
"eval_steps_per_second": 2.718, |
|
"step": 147 |
|
}, |
|
{ |
|
"epoch": 6.122448979591836, |
|
"grad_norm": 24.809764862060547, |
|
"learning_rate": 4.3981481481481486e-05, |
|
"loss": 0.8674, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 6.530612244897959, |
|
"grad_norm": 41.678409576416016, |
|
"learning_rate": 4.3209876543209875e-05, |
|
"loss": 0.8796, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 6.938775510204081, |
|
"grad_norm": 37.33673095703125, |
|
"learning_rate": 4.243827160493827e-05, |
|
"loss": 0.7568, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 6.979591836734694, |
|
"eval_accuracy": 0.45454545454545453, |
|
"eval_loss": 1.8934894800186157, |
|
"eval_runtime": 3.9506, |
|
"eval_samples_per_second": 5.569, |
|
"eval_steps_per_second": 2.784, |
|
"step": 171 |
|
}, |
|
{ |
|
"epoch": 7.346938775510204, |
|
"grad_norm": 22.92353057861328, |
|
"learning_rate": 4.166666666666667e-05, |
|
"loss": 0.6483, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 7.755102040816326, |
|
"grad_norm": 16.21958351135254, |
|
"learning_rate": 4.089506172839506e-05, |
|
"loss": 0.4024, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_accuracy": 0.45454545454545453, |
|
"eval_loss": 1.6766648292541504, |
|
"eval_runtime": 4.0618, |
|
"eval_samples_per_second": 5.416, |
|
"eval_steps_per_second": 2.708, |
|
"step": 196 |
|
}, |
|
{ |
|
"epoch": 8.16326530612245, |
|
"grad_norm": 22.85525894165039, |
|
"learning_rate": 4.012345679012346e-05, |
|
"loss": 0.6058, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 8.571428571428571, |
|
"grad_norm": 13.849242210388184, |
|
"learning_rate": 3.935185185185186e-05, |
|
"loss": 0.4311, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 8.979591836734693, |
|
"grad_norm": 23.329771041870117, |
|
"learning_rate": 3.8580246913580246e-05, |
|
"loss": 0.814, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 8.979591836734693, |
|
"eval_accuracy": 0.36363636363636365, |
|
"eval_loss": 1.7112053632736206, |
|
"eval_runtime": 3.9427, |
|
"eval_samples_per_second": 5.58, |
|
"eval_steps_per_second": 2.79, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 9.387755102040817, |
|
"grad_norm": 24.695878982543945, |
|
"learning_rate": 3.780864197530865e-05, |
|
"loss": 0.561, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 9.795918367346939, |
|
"grad_norm": 21.95734405517578, |
|
"learning_rate": 3.7037037037037037e-05, |
|
"loss": 0.4346, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"eval_accuracy": 0.4090909090909091, |
|
"eval_loss": 1.9364336729049683, |
|
"eval_runtime": 4.1094, |
|
"eval_samples_per_second": 5.354, |
|
"eval_steps_per_second": 2.677, |
|
"step": 245 |
|
}, |
|
{ |
|
"epoch": 10.204081632653061, |
|
"grad_norm": 9.583707809448242, |
|
"learning_rate": 3.626543209876543e-05, |
|
"loss": 0.316, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 10.612244897959183, |
|
"grad_norm": 3.692790985107422, |
|
"learning_rate": 3.5493827160493834e-05, |
|
"loss": 0.3456, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 10.979591836734693, |
|
"eval_accuracy": 0.5454545454545454, |
|
"eval_loss": 1.9417484998703003, |
|
"eval_runtime": 3.9243, |
|
"eval_samples_per_second": 5.606, |
|
"eval_steps_per_second": 2.803, |
|
"step": 269 |
|
}, |
|
{ |
|
"epoch": 11.020408163265307, |
|
"grad_norm": 20.5589599609375, |
|
"learning_rate": 3.472222222222222e-05, |
|
"loss": 0.4472, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 11.428571428571429, |
|
"grad_norm": 25.139612197875977, |
|
"learning_rate": 3.395061728395062e-05, |
|
"loss": 0.5478, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 11.83673469387755, |
|
"grad_norm": 21.445762634277344, |
|
"learning_rate": 3.317901234567901e-05, |
|
"loss": 0.228, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 12.0, |
|
"eval_accuracy": 0.4090909090909091, |
|
"eval_loss": 2.156919240951538, |
|
"eval_runtime": 4.1142, |
|
"eval_samples_per_second": 5.347, |
|
"eval_steps_per_second": 2.674, |
|
"step": 294 |
|
}, |
|
{ |
|
"epoch": 12.244897959183673, |
|
"grad_norm": 0.8465752601623535, |
|
"learning_rate": 3.240740740740741e-05, |
|
"loss": 0.1284, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 12.653061224489797, |
|
"grad_norm": 1.3978017568588257, |
|
"learning_rate": 3.16358024691358e-05, |
|
"loss": 0.1681, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 12.979591836734693, |
|
"eval_accuracy": 0.45454545454545453, |
|
"eval_loss": 2.056508779525757, |
|
"eval_runtime": 3.9483, |
|
"eval_samples_per_second": 5.572, |
|
"eval_steps_per_second": 2.786, |
|
"step": 318 |
|
}, |
|
{ |
|
"epoch": 13.061224489795919, |
|
"grad_norm": 0.8498027920722961, |
|
"learning_rate": 3.08641975308642e-05, |
|
"loss": 0.396, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 13.46938775510204, |
|
"grad_norm": 15.11242961883545, |
|
"learning_rate": 3.0092592592592593e-05, |
|
"loss": 0.2593, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 13.877551020408163, |
|
"grad_norm": 9.556538581848145, |
|
"learning_rate": 2.9320987654320992e-05, |
|
"loss": 0.1498, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 14.0, |
|
"eval_accuracy": 0.36363636363636365, |
|
"eval_loss": 2.070053815841675, |
|
"eval_runtime": 4.0317, |
|
"eval_samples_per_second": 5.457, |
|
"eval_steps_per_second": 2.728, |
|
"step": 343 |
|
}, |
|
{ |
|
"epoch": 14.285714285714286, |
|
"grad_norm": 5.47396993637085, |
|
"learning_rate": 2.8549382716049384e-05, |
|
"loss": 0.1807, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 14.693877551020408, |
|
"grad_norm": 21.753263473510742, |
|
"learning_rate": 2.777777777777778e-05, |
|
"loss": 0.1599, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 14.979591836734693, |
|
"eval_accuracy": 0.5, |
|
"eval_loss": 2.4973413944244385, |
|
"eval_runtime": 4.128, |
|
"eval_samples_per_second": 5.329, |
|
"eval_steps_per_second": 2.665, |
|
"step": 367 |
|
}, |
|
{ |
|
"epoch": 15.10204081632653, |
|
"grad_norm": 38.01487350463867, |
|
"learning_rate": 2.700617283950617e-05, |
|
"loss": 0.2408, |
|
"step": 370 |
|
}, |
|
{ |
|
"epoch": 15.510204081632653, |
|
"grad_norm": 13.578157424926758, |
|
"learning_rate": 2.623456790123457e-05, |
|
"loss": 0.2606, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 15.918367346938776, |
|
"grad_norm": 7.735710620880127, |
|
"learning_rate": 2.5462962962962965e-05, |
|
"loss": 0.3856, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 16.0, |
|
"eval_accuracy": 0.45454545454545453, |
|
"eval_loss": 2.247347831726074, |
|
"eval_runtime": 4.1251, |
|
"eval_samples_per_second": 5.333, |
|
"eval_steps_per_second": 2.667, |
|
"step": 392 |
|
}, |
|
{ |
|
"epoch": 16.3265306122449, |
|
"grad_norm": 6.533689975738525, |
|
"learning_rate": 2.4691358024691357e-05, |
|
"loss": 0.1284, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 16.73469387755102, |
|
"grad_norm": 22.71470069885254, |
|
"learning_rate": 2.3919753086419755e-05, |
|
"loss": 0.2529, |
|
"step": 410 |
|
}, |
|
{ |
|
"epoch": 16.979591836734695, |
|
"eval_accuracy": 0.45454545454545453, |
|
"eval_loss": 2.0918023586273193, |
|
"eval_runtime": 3.917, |
|
"eval_samples_per_second": 5.617, |
|
"eval_steps_per_second": 2.808, |
|
"step": 416 |
|
}, |
|
{ |
|
"epoch": 17.142857142857142, |
|
"grad_norm": 16.790376663208008, |
|
"learning_rate": 2.314814814814815e-05, |
|
"loss": 0.3629, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 17.551020408163264, |
|
"grad_norm": 14.542320251464844, |
|
"learning_rate": 2.2376543209876542e-05, |
|
"loss": 0.15, |
|
"step": 430 |
|
}, |
|
{ |
|
"epoch": 17.959183673469386, |
|
"grad_norm": 2.7366743087768555, |
|
"learning_rate": 2.1604938271604937e-05, |
|
"loss": 0.0557, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 18.0, |
|
"eval_accuracy": 0.5454545454545454, |
|
"eval_loss": 1.9595649242401123, |
|
"eval_runtime": 4.2837, |
|
"eval_samples_per_second": 5.136, |
|
"eval_steps_per_second": 2.568, |
|
"step": 441 |
|
}, |
|
{ |
|
"epoch": 18.367346938775512, |
|
"grad_norm": 51.391292572021484, |
|
"learning_rate": 2.0833333333333336e-05, |
|
"loss": 0.2554, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 18.775510204081634, |
|
"grad_norm": 18.502050399780273, |
|
"learning_rate": 2.006172839506173e-05, |
|
"loss": 0.0895, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 18.979591836734695, |
|
"eval_accuracy": 0.45454545454545453, |
|
"eval_loss": 2.5522444248199463, |
|
"eval_runtime": 4.1985, |
|
"eval_samples_per_second": 5.24, |
|
"eval_steps_per_second": 2.62, |
|
"step": 465 |
|
}, |
|
{ |
|
"epoch": 19.183673469387756, |
|
"grad_norm": 0.6463891267776489, |
|
"learning_rate": 1.9290123456790123e-05, |
|
"loss": 0.1512, |
|
"step": 470 |
|
}, |
|
{ |
|
"epoch": 19.591836734693878, |
|
"grad_norm": 2.839521646499634, |
|
"learning_rate": 1.8518518518518518e-05, |
|
"loss": 0.1938, |
|
"step": 480 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"grad_norm": 0.6800824403762817, |
|
"learning_rate": 1.7746913580246917e-05, |
|
"loss": 0.0719, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"eval_accuracy": 0.5, |
|
"eval_loss": 2.2938432693481445, |
|
"eval_runtime": 4.0864, |
|
"eval_samples_per_second": 5.384, |
|
"eval_steps_per_second": 2.692, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 20.408163265306122, |
|
"grad_norm": 14.73177719116211, |
|
"learning_rate": 1.697530864197531e-05, |
|
"loss": 0.2486, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 20.816326530612244, |
|
"grad_norm": 1.2800395488739014, |
|
"learning_rate": 1.6203703703703704e-05, |
|
"loss": 0.0764, |
|
"step": 510 |
|
}, |
|
{ |
|
"epoch": 20.979591836734695, |
|
"eval_accuracy": 0.45454545454545453, |
|
"eval_loss": 2.6754062175750732, |
|
"eval_runtime": 3.9516, |
|
"eval_samples_per_second": 5.567, |
|
"eval_steps_per_second": 2.784, |
|
"step": 514 |
|
}, |
|
{ |
|
"epoch": 21.224489795918366, |
|
"grad_norm": 33.9787483215332, |
|
"learning_rate": 1.54320987654321e-05, |
|
"loss": 0.2982, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 21.632653061224488, |
|
"grad_norm": 0.02819265052676201, |
|
"learning_rate": 1.4660493827160496e-05, |
|
"loss": 0.1301, |
|
"step": 530 |
|
}, |
|
{ |
|
"epoch": 22.0, |
|
"eval_accuracy": 0.45454545454545453, |
|
"eval_loss": 2.5286881923675537, |
|
"eval_runtime": 4.1246, |
|
"eval_samples_per_second": 5.334, |
|
"eval_steps_per_second": 2.667, |
|
"step": 539 |
|
}, |
|
{ |
|
"epoch": 22.040816326530614, |
|
"grad_norm": 21.637840270996094, |
|
"learning_rate": 1.388888888888889e-05, |
|
"loss": 0.0917, |
|
"step": 540 |
|
}, |
|
{ |
|
"epoch": 22.448979591836736, |
|
"grad_norm": 0.41712021827697754, |
|
"learning_rate": 1.3117283950617285e-05, |
|
"loss": 0.1209, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 22.857142857142858, |
|
"grad_norm": 26.01407241821289, |
|
"learning_rate": 1.2345679012345678e-05, |
|
"loss": 0.1205, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 22.979591836734695, |
|
"eval_accuracy": 0.4090909090909091, |
|
"eval_loss": 2.753232717514038, |
|
"eval_runtime": 3.9124, |
|
"eval_samples_per_second": 5.623, |
|
"eval_steps_per_second": 2.812, |
|
"step": 563 |
|
}, |
|
{ |
|
"epoch": 23.26530612244898, |
|
"grad_norm": 11.909204483032227, |
|
"learning_rate": 1.1574074074074075e-05, |
|
"loss": 0.1367, |
|
"step": 570 |
|
}, |
|
{ |
|
"epoch": 23.6734693877551, |
|
"grad_norm": 0.861128032207489, |
|
"learning_rate": 1.0802469135802469e-05, |
|
"loss": 0.1013, |
|
"step": 580 |
|
}, |
|
{ |
|
"epoch": 24.0, |
|
"eval_accuracy": 0.45454545454545453, |
|
"eval_loss": 2.6988065242767334, |
|
"eval_runtime": 4.1089, |
|
"eval_samples_per_second": 5.354, |
|
"eval_steps_per_second": 2.677, |
|
"step": 588 |
|
}, |
|
{ |
|
"epoch": 24.081632653061224, |
|
"grad_norm": 0.21600103378295898, |
|
"learning_rate": 1.0030864197530866e-05, |
|
"loss": 0.1176, |
|
"step": 590 |
|
}, |
|
{ |
|
"epoch": 24.489795918367346, |
|
"grad_norm": 39.00993347167969, |
|
"learning_rate": 9.259259259259259e-06, |
|
"loss": 0.1305, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 24.897959183673468, |
|
"grad_norm": 23.242216110229492, |
|
"learning_rate": 8.487654320987654e-06, |
|
"loss": 0.0777, |
|
"step": 610 |
|
}, |
|
{ |
|
"epoch": 24.979591836734695, |
|
"eval_accuracy": 0.4090909090909091, |
|
"eval_loss": 2.934515953063965, |
|
"eval_runtime": 3.9343, |
|
"eval_samples_per_second": 5.592, |
|
"eval_steps_per_second": 2.796, |
|
"step": 612 |
|
}, |
|
{ |
|
"epoch": 25.306122448979593, |
|
"grad_norm": 0.09061746299266815, |
|
"learning_rate": 7.71604938271605e-06, |
|
"loss": 0.0301, |
|
"step": 620 |
|
}, |
|
{ |
|
"epoch": 25.714285714285715, |
|
"grad_norm": 0.2142326533794403, |
|
"learning_rate": 6.944444444444445e-06, |
|
"loss": 0.1807, |
|
"step": 630 |
|
}, |
|
{ |
|
"epoch": 26.0, |
|
"eval_accuracy": 0.45454545454545453, |
|
"eval_loss": 2.998149871826172, |
|
"eval_runtime": 4.0792, |
|
"eval_samples_per_second": 5.393, |
|
"eval_steps_per_second": 2.697, |
|
"step": 637 |
|
}, |
|
{ |
|
"epoch": 26.122448979591837, |
|
"grad_norm": 0.43653419613838196, |
|
"learning_rate": 6.172839506172839e-06, |
|
"loss": 0.0358, |
|
"step": 640 |
|
}, |
|
{ |
|
"epoch": 26.53061224489796, |
|
"grad_norm": 0.5854562520980835, |
|
"learning_rate": 5.401234567901234e-06, |
|
"loss": 0.0422, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 26.93877551020408, |
|
"grad_norm": 0.39157870411872864, |
|
"learning_rate": 4.6296296296296296e-06, |
|
"loss": 0.0298, |
|
"step": 660 |
|
}, |
|
{ |
|
"epoch": 26.979591836734695, |
|
"eval_accuracy": 0.45454545454545453, |
|
"eval_loss": 2.854871988296509, |
|
"eval_runtime": 3.9772, |
|
"eval_samples_per_second": 5.532, |
|
"eval_steps_per_second": 2.766, |
|
"step": 661 |
|
}, |
|
{ |
|
"epoch": 27.346938775510203, |
|
"grad_norm": 0.041275933384895325, |
|
"learning_rate": 3.858024691358025e-06, |
|
"loss": 0.0633, |
|
"step": 670 |
|
}, |
|
{ |
|
"epoch": 27.755102040816325, |
|
"grad_norm": 18.42709732055664, |
|
"learning_rate": 3.0864197530864196e-06, |
|
"loss": 0.0589, |
|
"step": 680 |
|
}, |
|
{ |
|
"epoch": 28.0, |
|
"eval_accuracy": 0.45454545454545453, |
|
"eval_loss": 2.6966726779937744, |
|
"eval_runtime": 4.1347, |
|
"eval_samples_per_second": 5.321, |
|
"eval_steps_per_second": 2.66, |
|
"step": 686 |
|
}, |
|
{ |
|
"epoch": 28.163265306122447, |
|
"grad_norm": 27.545507431030273, |
|
"learning_rate": 2.3148148148148148e-06, |
|
"loss": 0.1035, |
|
"step": 690 |
|
}, |
|
{ |
|
"epoch": 28.571428571428573, |
|
"grad_norm": 14.813511848449707, |
|
"learning_rate": 1.5432098765432098e-06, |
|
"loss": 0.1094, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 28.979591836734695, |
|
"grad_norm": 0.47487640380859375, |
|
"learning_rate": 7.716049382716049e-07, |
|
"loss": 0.0896, |
|
"step": 710 |
|
}, |
|
{ |
|
"epoch": 28.979591836734695, |
|
"eval_accuracy": 0.45454545454545453, |
|
"eval_loss": 2.6902878284454346, |
|
"eval_runtime": 3.9324, |
|
"eval_samples_per_second": 5.595, |
|
"eval_steps_per_second": 2.797, |
|
"step": 710 |
|
}, |
|
{ |
|
"epoch": 29.387755102040817, |
|
"grad_norm": 0.2644538879394531, |
|
"learning_rate": 0.0, |
|
"loss": 0.0218, |
|
"step": 720 |
|
}, |
|
{ |
|
"epoch": 29.387755102040817, |
|
"eval_accuracy": 0.45454545454545453, |
|
"eval_loss": 2.690171480178833, |
|
"eval_runtime": 4.1046, |
|
"eval_samples_per_second": 5.36, |
|
"eval_steps_per_second": 2.68, |
|
"step": 720 |
|
}, |
|
{ |
|
"epoch": 29.387755102040817, |
|
"step": 720, |
|
"total_flos": 2.5282019639994163e+17, |
|
"train_loss": 0.5089964731906851, |
|
"train_runtime": 1181.0861, |
|
"train_samples_per_second": 4.953, |
|
"train_steps_per_second": 0.61 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 720, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 30, |
|
"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": 2.5282019639994163e+17, |
|
"train_batch_size": 2, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|