| { | |
| "best_metric": 0.894109396914446, | |
| "best_model_checkpoint": "swin-tiny-patch4-window7-224-classification/checkpoint-1000", | |
| "epoch": 9.975062344139651, | |
| "eval_steps": 500, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 5.313328742980957, | |
| "learning_rate": 1e-05, | |
| "loss": 2.0631, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 5.775210857391357, | |
| "learning_rate": 2e-05, | |
| "loss": 1.7746, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 4.816771984100342, | |
| "learning_rate": 3e-05, | |
| "loss": 1.5563, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 6.007688999176025, | |
| "learning_rate": 4e-05, | |
| "loss": 1.2361, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 7.807107925415039, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0685, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 6.920202732086182, | |
| "learning_rate": 6e-05, | |
| "loss": 0.9716, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 5.678104877471924, | |
| "learning_rate": 7e-05, | |
| "loss": 0.9274, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 7.1128010749816895, | |
| "learning_rate": 8e-05, | |
| "loss": 0.8962, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 8.700326919555664, | |
| "learning_rate": 9e-05, | |
| "loss": 0.8924, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 7.496997833251953, | |
| "learning_rate": 0.0001, | |
| "loss": 0.8514, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.705820476858345, | |
| "eval_loss": 0.7862311005592346, | |
| "eval_runtime": 38.0587, | |
| "eval_samples_per_second": 74.937, | |
| "eval_steps_per_second": 1.182, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 5.842683792114258, | |
| "learning_rate": 9.888888888888889e-05, | |
| "loss": 0.7961, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 7.082647323608398, | |
| "learning_rate": 9.777777777777778e-05, | |
| "loss": 0.8179, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 6.531412601470947, | |
| "learning_rate": 9.666666666666667e-05, | |
| "loss": 0.772, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 4.981847763061523, | |
| "learning_rate": 9.555555555555557e-05, | |
| "loss": 0.7298, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 6.9995198249816895, | |
| "learning_rate": 9.444444444444444e-05, | |
| "loss": 0.6882, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 4.9766693115234375, | |
| "learning_rate": 9.333333333333334e-05, | |
| "loss": 0.7048, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 5.899988174438477, | |
| "learning_rate": 9.222222222222223e-05, | |
| "loss": 0.7254, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 4.040374755859375, | |
| "learning_rate": 9.111111111111112e-05, | |
| "loss": 0.6543, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 6.156548023223877, | |
| "learning_rate": 9e-05, | |
| "loss": 0.6329, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 6.268504619598389, | |
| "learning_rate": 8.888888888888889e-05, | |
| "loss": 0.631, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7885694249649369, | |
| "eval_loss": 0.5491923093795776, | |
| "eval_runtime": 38.032, | |
| "eval_samples_per_second": 74.989, | |
| "eval_steps_per_second": 1.183, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "grad_norm": 6.425785541534424, | |
| "learning_rate": 8.777777777777778e-05, | |
| "loss": 0.612, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "grad_norm": 5.754735469818115, | |
| "learning_rate": 8.666666666666667e-05, | |
| "loss": 0.6117, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "grad_norm": 5.51466703414917, | |
| "learning_rate": 8.555555555555556e-05, | |
| "loss": 0.6041, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "grad_norm": 5.170792102813721, | |
| "learning_rate": 8.444444444444444e-05, | |
| "loss": 0.6057, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "grad_norm": 5.319341659545898, | |
| "learning_rate": 8.333333333333334e-05, | |
| "loss": 0.556, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "grad_norm": 7.033453941345215, | |
| "learning_rate": 8.222222222222222e-05, | |
| "loss": 0.5918, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "grad_norm": 6.37441349029541, | |
| "learning_rate": 8.111111111111112e-05, | |
| "loss": 0.5603, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "grad_norm": 4.926444053649902, | |
| "learning_rate": 8e-05, | |
| "loss": 0.5827, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "grad_norm": 4.273934841156006, | |
| "learning_rate": 7.88888888888889e-05, | |
| "loss": 0.5406, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "grad_norm": 5.528937816619873, | |
| "learning_rate": 7.777777777777778e-05, | |
| "loss": 0.5516, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 0.8078541374474053, | |
| "eval_loss": 0.5194684863090515, | |
| "eval_runtime": 38.7637, | |
| "eval_samples_per_second": 73.574, | |
| "eval_steps_per_second": 1.161, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "grad_norm": 9.39556884765625, | |
| "learning_rate": 7.666666666666667e-05, | |
| "loss": 0.5039, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "grad_norm": 5.658567428588867, | |
| "learning_rate": 7.555555555555556e-05, | |
| "loss": 0.497, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "grad_norm": 5.974184513092041, | |
| "learning_rate": 7.444444444444444e-05, | |
| "loss": 0.5041, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "grad_norm": 5.876557350158691, | |
| "learning_rate": 7.333333333333333e-05, | |
| "loss": 0.4978, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "grad_norm": 7.005821704864502, | |
| "learning_rate": 7.222222222222222e-05, | |
| "loss": 0.5145, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "grad_norm": 4.967439651489258, | |
| "learning_rate": 7.111111111111112e-05, | |
| "loss": 0.4803, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "grad_norm": 4.976831436157227, | |
| "learning_rate": 7e-05, | |
| "loss": 0.4816, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "grad_norm": 6.872544288635254, | |
| "learning_rate": 6.88888888888889e-05, | |
| "loss": 0.4811, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "grad_norm": 5.97269868850708, | |
| "learning_rate": 6.777777777777778e-05, | |
| "loss": 0.4978, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "grad_norm": 5.982359886169434, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.4597, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8229312762973352, | |
| "eval_loss": 0.48398053646087646, | |
| "eval_runtime": 38.5586, | |
| "eval_samples_per_second": 73.965, | |
| "eval_steps_per_second": 1.167, | |
| "step": 401 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "grad_norm": 4.751348972320557, | |
| "learning_rate": 6.555555555555556e-05, | |
| "loss": 0.4443, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "grad_norm": 4.549768447875977, | |
| "learning_rate": 6.444444444444446e-05, | |
| "loss": 0.423, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 4.29, | |
| "grad_norm": 6.269441604614258, | |
| "learning_rate": 6.333333333333333e-05, | |
| "loss": 0.449, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "grad_norm": 4.581325531005859, | |
| "learning_rate": 6.222222222222222e-05, | |
| "loss": 0.4099, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "grad_norm": 5.31665563583374, | |
| "learning_rate": 6.111111111111112e-05, | |
| "loss": 0.4075, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "grad_norm": 5.4510416984558105, | |
| "learning_rate": 6e-05, | |
| "loss": 0.4195, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "grad_norm": 5.312283515930176, | |
| "learning_rate": 5.8888888888888896e-05, | |
| "loss": 0.3829, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "grad_norm": 5.865201950073242, | |
| "learning_rate": 5.7777777777777776e-05, | |
| "loss": 0.4209, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "grad_norm": 5.123416423797607, | |
| "learning_rate": 5.666666666666667e-05, | |
| "loss": 0.4094, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "grad_norm": 4.805253982543945, | |
| "learning_rate": 5.555555555555556e-05, | |
| "loss": 0.4094, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8597475455820477, | |
| "eval_loss": 0.3945947587490082, | |
| "eval_runtime": 38.9464, | |
| "eval_samples_per_second": 73.229, | |
| "eval_steps_per_second": 1.155, | |
| "step": 501 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "grad_norm": 4.796308517456055, | |
| "learning_rate": 5.4444444444444446e-05, | |
| "loss": 0.3555, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "grad_norm": 4.460272789001465, | |
| "learning_rate": 5.333333333333333e-05, | |
| "loss": 0.3719, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 5.29, | |
| "grad_norm": 5.035310745239258, | |
| "learning_rate": 5.222222222222223e-05, | |
| "loss": 0.3785, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 5.39, | |
| "grad_norm": 5.145148754119873, | |
| "learning_rate": 5.111111111111111e-05, | |
| "loss": 0.3881, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "grad_norm": 6.646022319793701, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3776, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "grad_norm": 6.039920330047607, | |
| "learning_rate": 4.888888888888889e-05, | |
| "loss": 0.3891, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "grad_norm": 4.1056623458862305, | |
| "learning_rate": 4.7777777777777784e-05, | |
| "loss": 0.3369, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "grad_norm": 5.633683204650879, | |
| "learning_rate": 4.666666666666667e-05, | |
| "loss": 0.3606, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 5.89, | |
| "grad_norm": 5.873307704925537, | |
| "learning_rate": 4.555555555555556e-05, | |
| "loss": 0.3386, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "grad_norm": 5.3684983253479, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.3289, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8572931276297335, | |
| "eval_loss": 0.3786664605140686, | |
| "eval_runtime": 40.3623, | |
| "eval_samples_per_second": 70.66, | |
| "eval_steps_per_second": 1.115, | |
| "step": 601 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "grad_norm": 4.093196392059326, | |
| "learning_rate": 4.3333333333333334e-05, | |
| "loss": 0.3251, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.18, | |
| "grad_norm": 4.739712715148926, | |
| "learning_rate": 4.222222222222222e-05, | |
| "loss": 0.3029, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "grad_norm": 3.9042539596557617, | |
| "learning_rate": 4.111111111111111e-05, | |
| "loss": 0.3269, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 6.38, | |
| "grad_norm": 5.695065498352051, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3318, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 6.48, | |
| "grad_norm": 4.814879894256592, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.3087, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 6.58, | |
| "grad_norm": 4.633285999298096, | |
| "learning_rate": 3.777777777777778e-05, | |
| "loss": 0.3416, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 6.68, | |
| "grad_norm": 4.862430095672607, | |
| "learning_rate": 3.6666666666666666e-05, | |
| "loss": 0.3063, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "grad_norm": 4.930345058441162, | |
| "learning_rate": 3.555555555555556e-05, | |
| "loss": 0.3225, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "grad_norm": 4.587200164794922, | |
| "learning_rate": 3.444444444444445e-05, | |
| "loss": 0.3139, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "grad_norm": 5.570033550262451, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.3145, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "eval_accuracy": 0.8772791023842917, | |
| "eval_loss": 0.3536257743835449, | |
| "eval_runtime": 42.7073, | |
| "eval_samples_per_second": 66.78, | |
| "eval_steps_per_second": 1.054, | |
| "step": 701 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "grad_norm": 5.562831401824951, | |
| "learning_rate": 3.222222222222223e-05, | |
| "loss": 0.2998, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 7.18, | |
| "grad_norm": 5.306057453155518, | |
| "learning_rate": 3.111111111111111e-05, | |
| "loss": 0.2807, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 7.28, | |
| "grad_norm": 5.671096324920654, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2794, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 7.38, | |
| "grad_norm": 5.412108421325684, | |
| "learning_rate": 2.8888888888888888e-05, | |
| "loss": 0.2814, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "grad_norm": 4.249139308929443, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.2531, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 7.58, | |
| "grad_norm": 5.714820861816406, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 0.2712, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 7.68, | |
| "grad_norm": 5.9426469802856445, | |
| "learning_rate": 2.5555555555555554e-05, | |
| "loss": 0.2589, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "grad_norm": 4.787610054016113, | |
| "learning_rate": 2.4444444444444445e-05, | |
| "loss": 0.2745, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 7.88, | |
| "grad_norm": 3.9021761417388916, | |
| "learning_rate": 2.3333333333333336e-05, | |
| "loss": 0.2728, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 7.98, | |
| "grad_norm": 5.153603553771973, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.2911, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.885343618513324, | |
| "eval_loss": 0.32486388087272644, | |
| "eval_runtime": 42.9219, | |
| "eval_samples_per_second": 66.446, | |
| "eval_steps_per_second": 1.048, | |
| "step": 802 | |
| }, | |
| { | |
| "epoch": 8.08, | |
| "grad_norm": 5.644151210784912, | |
| "learning_rate": 2.111111111111111e-05, | |
| "loss": 0.2559, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 8.18, | |
| "grad_norm": 4.023806095123291, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2415, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "grad_norm": 6.214153289794922, | |
| "learning_rate": 1.888888888888889e-05, | |
| "loss": 0.2447, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 8.38, | |
| "grad_norm": 3.903085231781006, | |
| "learning_rate": 1.777777777777778e-05, | |
| "loss": 0.2602, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "grad_norm": 5.1828131675720215, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.2529, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 8.58, | |
| "grad_norm": 4.840915203094482, | |
| "learning_rate": 1.5555555555555555e-05, | |
| "loss": 0.2302, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 8.68, | |
| "grad_norm": 5.424919128417969, | |
| "learning_rate": 1.4444444444444444e-05, | |
| "loss": 0.246, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "grad_norm": 6.8232316970825195, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.2435, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 8.88, | |
| "grad_norm": 4.856207847595215, | |
| "learning_rate": 1.2222222222222222e-05, | |
| "loss": 0.2376, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "grad_norm": 5.239475250244141, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.2487, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8937587657784011, | |
| "eval_loss": 0.30939263105392456, | |
| "eval_runtime": 39.5102, | |
| "eval_samples_per_second": 72.184, | |
| "eval_steps_per_second": 1.139, | |
| "step": 902 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "grad_norm": 5.607595443725586, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2253, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 9.18, | |
| "grad_norm": 3.9586544036865234, | |
| "learning_rate": 8.88888888888889e-06, | |
| "loss": 0.2478, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 9.28, | |
| "grad_norm": 3.8814537525177, | |
| "learning_rate": 7.777777777777777e-06, | |
| "loss": 0.2173, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "grad_norm": 7.456895351409912, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.2315, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 9.48, | |
| "grad_norm": 4.543576717376709, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.2281, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 9.58, | |
| "grad_norm": 4.659777641296387, | |
| "learning_rate": 4.444444444444445e-06, | |
| "loss": 0.2324, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "grad_norm": 4.048426151275635, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.2132, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 9.78, | |
| "grad_norm": 3.969958543777466, | |
| "learning_rate": 2.2222222222222225e-06, | |
| "loss": 0.2076, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 9.88, | |
| "grad_norm": 5.621995449066162, | |
| "learning_rate": 1.1111111111111112e-06, | |
| "loss": 0.2262, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 9.98, | |
| "grad_norm": 4.668450355529785, | |
| "learning_rate": 0.0, | |
| "loss": 0.2291, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.98, | |
| "eval_accuracy": 0.894109396914446, | |
| "eval_loss": 0.303189218044281, | |
| "eval_runtime": 39.0907, | |
| "eval_samples_per_second": 72.958, | |
| "eval_steps_per_second": 1.151, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.98, | |
| "step": 1000, | |
| "total_flos": 6.364199987970048e+18, | |
| "train_loss": 0.48602974390983583, | |
| "train_runtime": 5351.8475, | |
| "train_samples_per_second": 47.954, | |
| "train_steps_per_second": 0.187 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 6.364199987970048e+18, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |