| { | |
| "best_metric": 0.08164331316947937, | |
| "best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-1100", | |
| "epoch": 4.0, | |
| "eval_steps": 100, | |
| "global_step": 1440, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 0.00019861111111111113, | |
| "loss": 1.5566, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 0.00019722222222222225, | |
| "loss": 1.262, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.00019583333333333334, | |
| "loss": 1.091, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00019444444444444446, | |
| "loss": 0.7922, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.00019305555555555558, | |
| "loss": 0.8375, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.00019166666666666667, | |
| "loss": 0.7567, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.00019027777777777776, | |
| "loss": 0.6015, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.00018888888888888888, | |
| "loss": 0.7216, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0001875, | |
| "loss": 0.6877, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.00018611111111111112, | |
| "loss": 0.5092, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "eval_accuracy": 0.7680555555555556, | |
| "eval_loss": 0.6419699788093567, | |
| "eval_runtime": 22.1712, | |
| "eval_samples_per_second": 32.475, | |
| "eval_steps_per_second": 4.059, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00018472222222222224, | |
| "loss": 0.4847, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 0.00018333333333333334, | |
| "loss": 0.6076, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00018194444444444445, | |
| "loss": 0.5853, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.00018055555555555557, | |
| "loss": 0.5133, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0001791666666666667, | |
| "loss": 0.4949, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00017777777777777779, | |
| "loss": 0.5219, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.0001763888888888889, | |
| "loss": 0.4862, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.000175, | |
| "loss": 0.5873, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.00017361111111111112, | |
| "loss": 0.4319, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00017222222222222224, | |
| "loss": 0.5076, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "eval_accuracy": 0.8722222222222222, | |
| "eval_loss": 0.4068766236305237, | |
| "eval_runtime": 22.7196, | |
| "eval_samples_per_second": 31.691, | |
| "eval_steps_per_second": 3.961, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.00017083333333333333, | |
| "loss": 0.4323, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 0.00016944444444444445, | |
| "loss": 0.361, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.00016805555555555557, | |
| "loss": 0.4156, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.0001666666666666667, | |
| "loss": 0.4297, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.00016527777777777778, | |
| "loss": 0.6522, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.0001638888888888889, | |
| "loss": 0.4157, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00016250000000000002, | |
| "loss": 0.3354, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.0001611111111111111, | |
| "loss": 0.3883, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.00015972222222222223, | |
| "loss": 0.4386, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 0.00015833333333333332, | |
| "loss": 0.3291, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "eval_accuracy": 0.8569444444444444, | |
| "eval_loss": 0.4342160224914551, | |
| "eval_runtime": 21.9024, | |
| "eval_samples_per_second": 32.873, | |
| "eval_steps_per_second": 4.109, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.00015694444444444444, | |
| "loss": 0.4767, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.00015555555555555556, | |
| "loss": 0.3492, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.00015416666666666668, | |
| "loss": 0.4947, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 0.00015277777777777777, | |
| "loss": 0.4482, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.0001513888888888889, | |
| "loss": 0.3367, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.2316, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 0.00014861111111111113, | |
| "loss": 0.2207, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 0.00014722222222222223, | |
| "loss": 0.1652, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.00014583333333333335, | |
| "loss": 0.2019, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 0.00014444444444444444, | |
| "loss": 0.108, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "eval_accuracy": 0.9291666666666667, | |
| "eval_loss": 0.24096721410751343, | |
| "eval_runtime": 21.4632, | |
| "eval_samples_per_second": 33.546, | |
| "eval_steps_per_second": 4.193, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 0.00014305555555555556, | |
| "loss": 0.1746, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.00014166666666666668, | |
| "loss": 0.0908, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 0.00014027777777777777, | |
| "loss": 0.0816, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 0.0001388888888888889, | |
| "loss": 0.1607, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 0.0001375, | |
| "loss": 0.0993, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 0.00013611111111111113, | |
| "loss": 0.1251, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.00013472222222222225, | |
| "loss": 0.1353, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 0.0739, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 0.00013194444444444446, | |
| "loss": 0.06, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 0.00013055555555555555, | |
| "loss": 0.0378, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "eval_accuracy": 0.9138888888888889, | |
| "eval_loss": 0.31065165996551514, | |
| "eval_runtime": 21.6753, | |
| "eval_samples_per_second": 33.217, | |
| "eval_steps_per_second": 4.152, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 0.00012916666666666667, | |
| "loss": 0.0966, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.00012777777777777776, | |
| "loss": 0.0644, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 0.00012638888888888888, | |
| "loss": 0.0791, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 0.000125, | |
| "loss": 0.1598, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.00012361111111111112, | |
| "loss": 0.1561, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.00012222222222222224, | |
| "loss": 0.1597, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 0.00012083333333333333, | |
| "loss": 0.2064, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 0.00011944444444444445, | |
| "loss": 0.1435, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 0.00011805555555555556, | |
| "loss": 0.0582, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.00011666666666666668, | |
| "loss": 0.1488, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "eval_accuracy": 0.9388888888888889, | |
| "eval_loss": 0.19844678044319153, | |
| "eval_runtime": 21.7891, | |
| "eval_samples_per_second": 33.044, | |
| "eval_steps_per_second": 4.131, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.00011527777777777777, | |
| "loss": 0.1309, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 0.00011388888888888889, | |
| "loss": 0.097, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.00011250000000000001, | |
| "loss": 0.1493, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 0.00011111111111111112, | |
| "loss": 0.1808, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 0.00010972222222222224, | |
| "loss": 0.1028, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 0.00010833333333333333, | |
| "loss": 0.096, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 0.00010694444444444445, | |
| "loss": 0.0943, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.00010555555555555557, | |
| "loss": 0.116, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 0.00010416666666666667, | |
| "loss": 0.1912, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 0.00010277777777777778, | |
| "loss": 0.0532, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_accuracy": 0.9513888888888888, | |
| "eval_loss": 0.17138923704624176, | |
| "eval_runtime": 23.0226, | |
| "eval_samples_per_second": 31.274, | |
| "eval_steps_per_second": 3.909, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 0.00010138888888888889, | |
| "loss": 0.08, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0874, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 9.861111111111112e-05, | |
| "loss": 0.0968, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 9.722222222222223e-05, | |
| "loss": 0.0607, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 9.583333333333334e-05, | |
| "loss": 0.0303, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 9.444444444444444e-05, | |
| "loss": 0.0113, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 9.305555555555556e-05, | |
| "loss": 0.1197, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 9.166666666666667e-05, | |
| "loss": 0.0256, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 9.027777777777779e-05, | |
| "loss": 0.0141, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 8.888888888888889e-05, | |
| "loss": 0.0122, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "eval_accuracy": 0.9611111111111111, | |
| "eval_loss": 0.1333748996257782, | |
| "eval_runtime": 22.6484, | |
| "eval_samples_per_second": 31.79, | |
| "eval_steps_per_second": 3.974, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 8.75e-05, | |
| "loss": 0.0222, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 8.611111111111112e-05, | |
| "loss": 0.0101, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 8.472222222222222e-05, | |
| "loss": 0.042, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 8.333333333333334e-05, | |
| "loss": 0.0232, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 8.194444444444445e-05, | |
| "loss": 0.1049, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 8.055555555555556e-05, | |
| "loss": 0.0191, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 7.916666666666666e-05, | |
| "loss": 0.0778, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 7.777777777777778e-05, | |
| "loss": 0.0533, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 7.638888888888889e-05, | |
| "loss": 0.0567, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.0529, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "eval_accuracy": 0.9652777777777778, | |
| "eval_loss": 0.11391060799360275, | |
| "eval_runtime": 21.5745, | |
| "eval_samples_per_second": 33.373, | |
| "eval_steps_per_second": 4.172, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 7.361111111111111e-05, | |
| "loss": 0.0184, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 7.222222222222222e-05, | |
| "loss": 0.0247, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 7.083333333333334e-05, | |
| "loss": 0.0075, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 6.944444444444444e-05, | |
| "loss": 0.0601, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 6.805555555555556e-05, | |
| "loss": 0.0394, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.0079, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 6.527777777777778e-05, | |
| "loss": 0.0113, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 6.388888888888888e-05, | |
| "loss": 0.0417, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.0658, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 6.111111111111112e-05, | |
| "loss": 0.0221, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "eval_accuracy": 0.9736111111111111, | |
| "eval_loss": 0.08747294545173645, | |
| "eval_runtime": 21.8071, | |
| "eval_samples_per_second": 33.017, | |
| "eval_steps_per_second": 4.127, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 5.972222222222223e-05, | |
| "loss": 0.0223, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 5.833333333333334e-05, | |
| "loss": 0.0361, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 5.6944444444444445e-05, | |
| "loss": 0.0065, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 5.555555555555556e-05, | |
| "loss": 0.0062, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 5.4166666666666664e-05, | |
| "loss": 0.0083, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 5.2777777777777784e-05, | |
| "loss": 0.0077, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 5.138888888888889e-05, | |
| "loss": 0.0123, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0266, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 0.0058, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.0052, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "eval_accuracy": 0.9819444444444444, | |
| "eval_loss": 0.08164331316947937, | |
| "eval_runtime": 21.3086, | |
| "eval_samples_per_second": 33.789, | |
| "eval_steps_per_second": 4.224, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 4.5833333333333334e-05, | |
| "loss": 0.0049, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.0053, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 4.305555555555556e-05, | |
| "loss": 0.0295, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.0044, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 4.027777777777778e-05, | |
| "loss": 0.0058, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.0052, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.0043, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.0045, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 0.0043, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.0045, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "eval_accuracy": 0.9791666666666666, | |
| "eval_loss": 0.08730719983577728, | |
| "eval_runtime": 21.963, | |
| "eval_samples_per_second": 32.782, | |
| "eval_steps_per_second": 4.098, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 3.194444444444444e-05, | |
| "loss": 0.0044, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.0042, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 2.916666666666667e-05, | |
| "loss": 0.0392, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.0042, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 2.6388888888888892e-05, | |
| "loss": 0.0045, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0043, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 2.361111111111111e-05, | |
| "loss": 0.0044, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.0041, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.0043, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.0113, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "eval_accuracy": 0.9833333333333333, | |
| "eval_loss": 0.08817147463560104, | |
| "eval_runtime": 22.6142, | |
| "eval_samples_per_second": 31.838, | |
| "eval_steps_per_second": 3.98, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 1.8055555555555555e-05, | |
| "loss": 0.0038, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.0041, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 1.527777777777778e-05, | |
| "loss": 0.0039, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.0037, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.0042, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.006, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 9.722222222222223e-06, | |
| "loss": 0.0036, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.004, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.0035, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0043, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "eval_accuracy": 0.9805555555555555, | |
| "eval_loss": 0.08652874827384949, | |
| "eval_runtime": 21.5822, | |
| "eval_samples_per_second": 33.361, | |
| "eval_steps_per_second": 4.17, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 0.0043, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.0041, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "learning_rate": 1.388888888888889e-06, | |
| "loss": 0.0164, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.0038, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 1440, | |
| <<<<<<< HEAD | |
| "total_flos": 6.311528820941783e+18, | |
| "train_loss": 0.2221095090903772, | |
| "train_runtime": 4823.2045, | |
| "train_samples_per_second": 4.777, | |
| "train_steps_per_second": 0.299 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9791666666666666, | |
| "eval_loss": 0.10265839099884033, | |
| "eval_runtime": 48.2725, | |
| "eval_samples_per_second": 14.915, | |
| "eval_steps_per_second": 1.864, | |
| ======= | |
| "total_flos": 1.7854794491756544e+18, | |
| "train_loss": 0.18370533613229378, | |
| "train_runtime": 2041.1421, | |
| "train_samples_per_second": 11.288, | |
| "train_steps_per_second": 0.705 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9819444444444444, | |
| "eval_loss": 0.08164331316947937, | |
| "eval_runtime": 32.0666, | |
| "eval_samples_per_second": 22.453, | |
| "eval_steps_per_second": 2.807, | |
| >>>>>>> parent of c5d9224 (Upload folder using huggingface_hub) | |
| "step": 1440 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1440, | |
| "num_train_epochs": 4, | |
| "save_steps": 100, | |
| "total_flos": 1.7854794491756544e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |