| { | |
| "best_metric": 0.1909320205450058, | |
| "best_model_checkpoint": "./vit-base-fullbody-classifier-6\\checkpoint-1100", | |
| "epoch": 10.0, | |
| "eval_steps": 100, | |
| "global_step": 2070, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 1.3746, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 1.3829, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 3e-06, | |
| "loss": 1.3757, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.3656, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 5e-06, | |
| "loss": 1.3448, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 6e-06, | |
| "loss": 1.3373, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 7.000000000000001e-06, | |
| "loss": 1.3232, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.2872, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 9e-06, | |
| "loss": 1.2638, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 1e-05, | |
| "loss": 1.25, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_accuracy": 0.5926829268292683, | |
| "eval_loss": 1.2367020845413208, | |
| "eval_runtime": 20.8647, | |
| "eval_samples_per_second": 19.65, | |
| "eval_steps_per_second": 2.492, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 1.1748, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 1.2e-05, | |
| "loss": 1.1491, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 1.3000000000000001e-05, | |
| "loss": 1.0737, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 1.0358, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.9602, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.9154, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 1.7000000000000003e-05, | |
| "loss": 0.8374, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.7678, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 1.9e-05, | |
| "loss": 0.7105, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6594, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_accuracy": 0.8731707317073171, | |
| "eval_loss": 0.64021235704422, | |
| "eval_runtime": 21.6651, | |
| "eval_samples_per_second": 18.924, | |
| "eval_steps_per_second": 2.4, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 2.1e-05, | |
| "loss": 0.6217, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 0.6355, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 2.3000000000000003e-05, | |
| "loss": 0.5103, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.4564, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.3947, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 0.3605, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 2.7000000000000002e-05, | |
| "loss": 0.3896, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 0.3193, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 2.9e-05, | |
| "loss": 0.334, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2512, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "eval_accuracy": 0.948780487804878, | |
| "eval_loss": 0.28558439016342163, | |
| "eval_runtime": 22.0969, | |
| "eval_samples_per_second": 18.555, | |
| "eval_steps_per_second": 2.353, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 3.1e-05, | |
| "loss": 0.257, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.2081, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 3.3e-05, | |
| "loss": 0.2643, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 0.271, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.2353, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.316, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.1995, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.3356, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.2513, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 4e-05, | |
| "loss": 0.2175, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "eval_accuracy": 0.8975609756097561, | |
| "eval_loss": 0.3272535502910614, | |
| "eval_runtime": 21.8078, | |
| "eval_samples_per_second": 18.801, | |
| "eval_steps_per_second": 2.384, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.281, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.2321, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 4.29e-05, | |
| "loss": 0.1141, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 4.39e-05, | |
| "loss": 0.1023, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 4.49e-05, | |
| "loss": 0.1081, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 4.5900000000000004e-05, | |
| "loss": 0.0969, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 4.69e-05, | |
| "loss": 0.0822, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 4.79e-05, | |
| "loss": 0.1229, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 4.89e-05, | |
| "loss": 0.1556, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 4.99e-05, | |
| "loss": 0.1313, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "eval_accuracy": 0.9317073170731708, | |
| "eval_loss": 0.229569673538208, | |
| "eval_runtime": 20.4104, | |
| "eval_samples_per_second": 20.088, | |
| "eval_steps_per_second": 2.548, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 4.9713375796178345e-05, | |
| "loss": 0.1338, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 4.939490445859873e-05, | |
| "loss": 0.1007, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 4.907643312101911e-05, | |
| "loss": 0.1014, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 4.875796178343949e-05, | |
| "loss": 0.1331, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 4.8439490445859877e-05, | |
| "loss": 0.1637, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 4.8121019108280254e-05, | |
| "loss": 0.1037, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 4.780254777070064e-05, | |
| "loss": 0.1222, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 4.7484076433121024e-05, | |
| "loss": 0.0777, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 4.71656050955414e-05, | |
| "loss": 0.1369, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 4.6847133757961786e-05, | |
| "loss": 0.0698, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "eval_accuracy": 0.9414634146341463, | |
| "eval_loss": 0.2208263874053955, | |
| "eval_runtime": 20.9683, | |
| "eval_samples_per_second": 19.553, | |
| "eval_steps_per_second": 2.48, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 4.652866242038217e-05, | |
| "loss": 0.1643, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.621019108280255e-05, | |
| "loss": 0.1105, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 4.589171974522293e-05, | |
| "loss": 0.0556, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 4.557324840764332e-05, | |
| "loss": 0.0482, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 4.5254777070063695e-05, | |
| "loss": 0.0553, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 4.493630573248408e-05, | |
| "loss": 0.0677, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 4.4617834394904464e-05, | |
| "loss": 0.0365, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "learning_rate": 4.429936305732484e-05, | |
| "loss": 0.0529, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 4.398089171974523e-05, | |
| "loss": 0.041, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 4.366242038216561e-05, | |
| "loss": 0.0553, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "eval_accuracy": 0.9512195121951219, | |
| "eval_loss": 0.1926226019859314, | |
| "eval_runtime": 21.809, | |
| "eval_samples_per_second": 18.8, | |
| "eval_steps_per_second": 2.384, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "learning_rate": 4.334394904458599e-05, | |
| "loss": 0.0574, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 4.3025477707006374e-05, | |
| "loss": 0.0681, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 4.270700636942676e-05, | |
| "loss": 0.0541, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 4.2388535031847136e-05, | |
| "loss": 0.0574, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 4.207006369426752e-05, | |
| "loss": 0.0447, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 4.1751592356687905e-05, | |
| "loss": 0.0461, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 4.143312101910828e-05, | |
| "loss": 0.0228, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 4.111464968152867e-05, | |
| "loss": 0.034, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 4.0796178343949045e-05, | |
| "loss": 0.0557, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 4.047770700636943e-05, | |
| "loss": 0.0239, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "eval_accuracy": 0.9390243902439024, | |
| "eval_loss": 0.19786834716796875, | |
| "eval_runtime": 21.3483, | |
| "eval_samples_per_second": 19.205, | |
| "eval_steps_per_second": 2.436, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 4.0159235668789814e-05, | |
| "loss": 0.0534, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 3.984076433121019e-05, | |
| "loss": 0.0521, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 3.952229299363058e-05, | |
| "loss": 0.0423, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 3.920382165605096e-05, | |
| "loss": 0.0197, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 3.888535031847134e-05, | |
| "loss": 0.0584, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "learning_rate": 3.8566878980891724e-05, | |
| "loss": 0.0695, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "learning_rate": 3.824840764331211e-05, | |
| "loss": 0.039, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 3.7929936305732486e-05, | |
| "loss": 0.0183, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "learning_rate": 3.761146496815287e-05, | |
| "loss": 0.0175, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 3.729299363057325e-05, | |
| "loss": 0.0386, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "eval_accuracy": 0.9365853658536586, | |
| "eval_loss": 0.22525091469287872, | |
| "eval_runtime": 21.4031, | |
| "eval_samples_per_second": 19.156, | |
| "eval_steps_per_second": 2.43, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "learning_rate": 3.697452229299363e-05, | |
| "loss": 0.0164, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 3.665605095541401e-05, | |
| "loss": 0.0161, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "learning_rate": 3.6337579617834395e-05, | |
| "loss": 0.0209, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "learning_rate": 3.601910828025477e-05, | |
| "loss": 0.0668, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 3.570063694267516e-05, | |
| "loss": 0.0697, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 4.64, | |
| "learning_rate": 3.538216560509554e-05, | |
| "loss": 0.0161, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "learning_rate": 3.506369426751592e-05, | |
| "loss": 0.0783, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 3.4745222929936305e-05, | |
| "loss": 0.0152, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "learning_rate": 3.442675159235669e-05, | |
| "loss": 0.0146, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 3.410828025477707e-05, | |
| "loss": 0.0143, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "eval_accuracy": 0.9463414634146341, | |
| "eval_loss": 0.19345715641975403, | |
| "eval_runtime": 21.1331, | |
| "eval_samples_per_second": 19.401, | |
| "eval_steps_per_second": 2.461, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "learning_rate": 3.378980891719745e-05, | |
| "loss": 0.0143, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "learning_rate": 3.3471337579617836e-05, | |
| "loss": 0.0136, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "learning_rate": 3.3152866242038214e-05, | |
| "loss": 0.0386, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 3.28343949044586e-05, | |
| "loss": 0.0336, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "learning_rate": 3.251592356687898e-05, | |
| "loss": 0.013, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "learning_rate": 3.219745222929936e-05, | |
| "loss": 0.0126, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "learning_rate": 3.1878980891719745e-05, | |
| "loss": 0.0128, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 3.156050955414013e-05, | |
| "loss": 0.0122, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 5.27, | |
| "learning_rate": 3.124203821656051e-05, | |
| "loss": 0.0136, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 5.31, | |
| "learning_rate": 3.092356687898089e-05, | |
| "loss": 0.0606, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 5.31, | |
| "eval_accuracy": 0.9536585365853658, | |
| "eval_loss": 0.1909320205450058, | |
| "eval_runtime": 20.2405, | |
| "eval_samples_per_second": 20.256, | |
| "eval_steps_per_second": 2.569, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "learning_rate": 3.060509554140128e-05, | |
| "loss": 0.0368, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 5.41, | |
| "learning_rate": 3.0286624203821655e-05, | |
| "loss": 0.0116, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "learning_rate": 2.996815286624204e-05, | |
| "loss": 0.0335, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 5.51, | |
| "learning_rate": 2.964968152866242e-05, | |
| "loss": 0.0211, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "learning_rate": 2.93312101910828e-05, | |
| "loss": 0.0114, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 2.9012738853503186e-05, | |
| "loss": 0.0119, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "learning_rate": 2.8694267515923567e-05, | |
| "loss": 0.0167, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "learning_rate": 2.837579617834395e-05, | |
| "loss": 0.0111, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "learning_rate": 2.8057324840764333e-05, | |
| "loss": 0.011, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "learning_rate": 2.7738853503184714e-05, | |
| "loss": 0.0105, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "eval_accuracy": 0.9390243902439024, | |
| "eval_loss": 0.24669073522090912, | |
| "eval_runtime": 20.6758, | |
| "eval_samples_per_second": 19.83, | |
| "eval_steps_per_second": 2.515, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 5.85, | |
| "learning_rate": 2.7420382165605095e-05, | |
| "loss": 0.0104, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 5.89, | |
| "learning_rate": 2.710191082802548e-05, | |
| "loss": 0.0106, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "learning_rate": 2.678343949044586e-05, | |
| "loss": 0.0101, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "learning_rate": 2.6464968152866242e-05, | |
| "loss": 0.0395, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 2.6146496815286624e-05, | |
| "loss": 0.0098, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "learning_rate": 2.5828025477707008e-05, | |
| "loss": 0.0099, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 6.14, | |
| "learning_rate": 2.550955414012739e-05, | |
| "loss": 0.0097, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 6.18, | |
| "learning_rate": 2.519108280254777e-05, | |
| "loss": 0.0095, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 6.23, | |
| "learning_rate": 2.4872611464968155e-05, | |
| "loss": 0.0094, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "learning_rate": 2.4554140127388536e-05, | |
| "loss": 0.0092, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "eval_accuracy": 0.9512195121951219, | |
| "eval_loss": 0.19856394827365875, | |
| "eval_runtime": 19.7104, | |
| "eval_samples_per_second": 20.801, | |
| "eval_steps_per_second": 2.638, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 6.33, | |
| "learning_rate": 2.4235668789808917e-05, | |
| "loss": 0.0091, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 6.38, | |
| "learning_rate": 2.3917197452229302e-05, | |
| "loss": 0.009, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "learning_rate": 2.3598726114649683e-05, | |
| "loss": 0.0089, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 6.47, | |
| "learning_rate": 2.3280254777070064e-05, | |
| "loss": 0.0092, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "learning_rate": 2.296178343949045e-05, | |
| "loss": 0.0087, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 6.57, | |
| "learning_rate": 2.264331210191083e-05, | |
| "loss": 0.0086, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 2.232484076433121e-05, | |
| "loss": 0.029, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 2.2006369426751592e-05, | |
| "loss": 0.0087, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 6.71, | |
| "learning_rate": 2.1687898089171977e-05, | |
| "loss": 0.0087, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 6.76, | |
| "learning_rate": 2.1369426751592358e-05, | |
| "loss": 0.0085, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 6.76, | |
| "eval_accuracy": 0.9512195121951219, | |
| "eval_loss": 0.20551267266273499, | |
| "eval_runtime": 21.7651, | |
| "eval_samples_per_second": 18.837, | |
| "eval_steps_per_second": 2.389, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 6.81, | |
| "learning_rate": 2.105095541401274e-05, | |
| "loss": 0.0085, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 6.86, | |
| "learning_rate": 2.0732484076433124e-05, | |
| "loss": 0.0084, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "learning_rate": 2.0414012738853505e-05, | |
| "loss": 0.0081, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "learning_rate": 2.0095541401273886e-05, | |
| "loss": 0.008, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 1.977707006369427e-05, | |
| "loss": 0.0112, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 1.9458598726114652e-05, | |
| "loss": 0.0079, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 1.9140127388535033e-05, | |
| "loss": 0.0078, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 7.15, | |
| "learning_rate": 1.8821656050955418e-05, | |
| "loss": 0.0078, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 1.8503184713375796e-05, | |
| "loss": 0.0076, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "learning_rate": 1.8184713375796177e-05, | |
| "loss": 0.0076, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "eval_accuracy": 0.9463414634146341, | |
| "eval_loss": 0.2132251262664795, | |
| "eval_runtime": 21.8557, | |
| "eval_samples_per_second": 18.759, | |
| "eval_steps_per_second": 2.379, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.29, | |
| "learning_rate": 1.786624203821656e-05, | |
| "loss": 0.0075, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 7.34, | |
| "learning_rate": 1.7547770700636942e-05, | |
| "loss": 0.0075, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 7.39, | |
| "learning_rate": 1.7229299363057324e-05, | |
| "loss": 0.0075, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 7.44, | |
| "learning_rate": 1.6910828025477708e-05, | |
| "loss": 0.0074, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 7.49, | |
| "learning_rate": 1.659235668789809e-05, | |
| "loss": 0.0073, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 7.54, | |
| "learning_rate": 1.627388535031847e-05, | |
| "loss": 0.0073, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 7.58, | |
| "learning_rate": 1.5955414012738855e-05, | |
| "loss": 0.0073, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 7.63, | |
| "learning_rate": 1.5636942675159236e-05, | |
| "loss": 0.0072, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 7.68, | |
| "learning_rate": 1.5318471337579618e-05, | |
| "loss": 0.0072, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 7.73, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.0071, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 7.73, | |
| "eval_accuracy": 0.9463414634146341, | |
| "eval_loss": 0.21747459471225739, | |
| "eval_runtime": 20.4277, | |
| "eval_samples_per_second": 20.071, | |
| "eval_steps_per_second": 2.546, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "learning_rate": 1.4681528662420382e-05, | |
| "loss": 0.007, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "learning_rate": 1.4363057324840764e-05, | |
| "loss": 0.007, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 7.87, | |
| "learning_rate": 1.4044585987261147e-05, | |
| "loss": 0.0069, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 7.92, | |
| "learning_rate": 1.3726114649681528e-05, | |
| "loss": 0.0069, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 7.97, | |
| "learning_rate": 1.3407643312101911e-05, | |
| "loss": 0.0171, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 1.3089171974522294e-05, | |
| "loss": 0.0068, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 1.2770700636942675e-05, | |
| "loss": 0.0068, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "learning_rate": 1.2452229299363058e-05, | |
| "loss": 0.0068, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 8.16, | |
| "learning_rate": 1.213375796178344e-05, | |
| "loss": 0.0067, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 8.21, | |
| "learning_rate": 1.1815286624203822e-05, | |
| "loss": 0.0067, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 8.21, | |
| "eval_accuracy": 0.948780487804878, | |
| "eval_loss": 0.2226085364818573, | |
| "eval_runtime": 21.3206, | |
| "eval_samples_per_second": 19.23, | |
| "eval_steps_per_second": 2.439, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 8.26, | |
| "learning_rate": 1.1496815286624205e-05, | |
| "loss": 0.0067, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 8.31, | |
| "learning_rate": 1.1178343949044586e-05, | |
| "loss": 0.0066, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 8.36, | |
| "learning_rate": 1.085987261146497e-05, | |
| "loss": 0.0066, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 8.41, | |
| "learning_rate": 1.054140127388535e-05, | |
| "loss": 0.0065, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 8.45, | |
| "learning_rate": 1.0222929936305733e-05, | |
| "loss": 0.0065, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "learning_rate": 9.904458598726116e-06, | |
| "loss": 0.0065, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "learning_rate": 9.585987261146497e-06, | |
| "loss": 0.0065, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 8.6, | |
| "learning_rate": 9.267515923566879e-06, | |
| "loss": 0.0064, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 8.65, | |
| "learning_rate": 8.949044585987261e-06, | |
| "loss": 0.0064, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "learning_rate": 8.630573248407643e-06, | |
| "loss": 0.0064, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "eval_accuracy": 0.9463414634146341, | |
| "eval_loss": 0.22321389615535736, | |
| "eval_runtime": 19.9115, | |
| "eval_samples_per_second": 20.591, | |
| "eval_steps_per_second": 2.612, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 8.74, | |
| "learning_rate": 8.312101910828025e-06, | |
| "loss": 0.0064, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 8.79, | |
| "learning_rate": 7.993630573248408e-06, | |
| "loss": 0.0106, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 8.84, | |
| "learning_rate": 7.67515923566879e-06, | |
| "loss": 0.0064, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "learning_rate": 7.356687898089172e-06, | |
| "loss": 0.0063, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 8.94, | |
| "learning_rate": 7.0382165605095544e-06, | |
| "loss": 0.0063, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "learning_rate": 6.7197452229299365e-06, | |
| "loss": 0.013, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 6.4012738853503185e-06, | |
| "loss": 0.0062, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "learning_rate": 6.082802547770701e-06, | |
| "loss": 0.0062, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 9.13, | |
| "learning_rate": 5.764331210191083e-06, | |
| "loss": 0.0062, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 9.18, | |
| "learning_rate": 5.445859872611465e-06, | |
| "loss": 0.0062, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 9.18, | |
| "eval_accuracy": 0.9463414634146341, | |
| "eval_loss": 0.2225855439901352, | |
| "eval_runtime": 18.1543, | |
| "eval_samples_per_second": 22.584, | |
| "eval_steps_per_second": 2.864, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 9.23, | |
| "learning_rate": 5.1273885350318474e-06, | |
| "loss": 0.0061, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 9.28, | |
| "learning_rate": 4.8089171974522295e-06, | |
| "loss": 0.0062, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 9.32, | |
| "learning_rate": 4.4904458598726115e-06, | |
| "loss": 0.0061, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 9.37, | |
| "learning_rate": 4.1719745222929935e-06, | |
| "loss": 0.0062, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 9.42, | |
| "learning_rate": 3.8535031847133755e-06, | |
| "loss": 0.0061, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "learning_rate": 3.535031847133758e-06, | |
| "loss": 0.0061, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 3.2165605095541404e-06, | |
| "loss": 0.0061, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "learning_rate": 2.8980891719745225e-06, | |
| "loss": 0.0061, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 9.61, | |
| "learning_rate": 2.579617834394905e-06, | |
| "loss": 0.0061, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 9.66, | |
| "learning_rate": 2.2611464968152865e-06, | |
| "loss": 0.0061, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 9.66, | |
| "eval_accuracy": 0.9463414634146341, | |
| "eval_loss": 0.2239292562007904, | |
| "eval_runtime": 17.7456, | |
| "eval_samples_per_second": 23.104, | |
| "eval_steps_per_second": 2.93, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 9.71, | |
| "learning_rate": 1.942675159235669e-06, | |
| "loss": 0.0061, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 9.76, | |
| "learning_rate": 1.624203821656051e-06, | |
| "loss": 0.0061, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 9.81, | |
| "learning_rate": 1.3057324840764332e-06, | |
| "loss": 0.0097, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 9.86, | |
| "learning_rate": 9.872611464968153e-07, | |
| "loss": 0.0061, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 9.9, | |
| "learning_rate": 6.687898089171975e-07, | |
| "loss": 0.0061, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 9.95, | |
| "learning_rate": 3.503184713375796e-07, | |
| "loss": 0.0061, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 3.184713375796178e-08, | |
| "loss": 0.0112, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 2070, | |
| "total_flos": 2.557281496670208e+18, | |
| "train_loss": 0.16907700259904354, | |
| "train_runtime": 2197.7206, | |
| "train_samples_per_second": 15.016, | |
| "train_steps_per_second": 0.942 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2070, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 100, | |
| "total_flos": 2.557281496670208e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |