| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 19.860529986053, | |
| "global_step": 1780, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 4.971910112359551e-05, | |
| "loss": 6.9134, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 4.943820224719101e-05, | |
| "loss": 2.1399, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.915730337078652e-05, | |
| "loss": 1.0034, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 4.8876404494382024e-05, | |
| "loss": 0.4659, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 4.859550561797753e-05, | |
| "loss": 0.2436, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 4.831460674157304e-05, | |
| "loss": 0.1612, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_accuracy": 0.9375623695908555, | |
| "eval_loss": 0.11446994543075562, | |
| "eval_runtime": 8.4911, | |
| "eval_samples_per_second": 71.133, | |
| "eval_steps_per_second": 23.79, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 4.803370786516854e-05, | |
| "loss": 0.1068, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 4.7752808988764046e-05, | |
| "loss": 0.0831, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.747191011235955e-05, | |
| "loss": 0.0751, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 4.719101123595506e-05, | |
| "loss": 0.0706, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 4.691011235955056e-05, | |
| "loss": 0.0676, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 4.662921348314607e-05, | |
| "loss": 0.0666, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "eval_accuracy": 0.9356021824496832, | |
| "eval_loss": 0.06282978504896164, | |
| "eval_runtime": 8.387, | |
| "eval_samples_per_second": 72.016, | |
| "eval_steps_per_second": 24.085, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 4.6348314606741575e-05, | |
| "loss": 0.0648, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 4.606741573033708e-05, | |
| "loss": 0.0632, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 4.578651685393259e-05, | |
| "loss": 0.0616, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 4.550561797752809e-05, | |
| "loss": 0.0609, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 4.52247191011236e-05, | |
| "loss": 0.0608, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 4.4943820224719104e-05, | |
| "loss": 0.0599, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "eval_accuracy": 0.9355179430022939, | |
| "eval_loss": 0.061063703149557114, | |
| "eval_runtime": 8.4253, | |
| "eval_samples_per_second": 71.689, | |
| "eval_steps_per_second": 23.975, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 4.4662921348314605e-05, | |
| "loss": 0.0595, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 4.438202247191011e-05, | |
| "loss": 0.0586, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 4.410112359550562e-05, | |
| "loss": 0.058, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 4.3820224719101126e-05, | |
| "loss": 0.0573, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 4.353932584269663e-05, | |
| "loss": 0.0567, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 4.3258426966292134e-05, | |
| "loss": 0.0563, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "eval_accuracy": 0.9352101450214486, | |
| "eval_loss": 0.06314567476511002, | |
| "eval_runtime": 8.3902, | |
| "eval_samples_per_second": 71.989, | |
| "eval_steps_per_second": 24.076, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 4.297752808988764e-05, | |
| "loss": 0.0555, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 4.269662921348315e-05, | |
| "loss": 0.0543, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 4.2415730337078655e-05, | |
| "loss": 0.0544, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 4.2134831460674156e-05, | |
| "loss": 0.0528, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 4.185393258426967e-05, | |
| "loss": 0.052, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 4.157303370786517e-05, | |
| "loss": 0.0512, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "eval_accuracy": 0.9347176682520962, | |
| "eval_loss": 0.06297876685857773, | |
| "eval_runtime": 8.3784, | |
| "eval_samples_per_second": 72.09, | |
| "eval_steps_per_second": 24.11, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 4.129213483146068e-05, | |
| "loss": 0.05, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 4.1011235955056184e-05, | |
| "loss": 0.0498, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 4.0730337078651685e-05, | |
| "loss": 0.0493, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 4.044943820224719e-05, | |
| "loss": 0.0484, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 4.01685393258427e-05, | |
| "loss": 0.048, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 3.98876404494382e-05, | |
| "loss": 0.0472, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "eval_accuracy": 0.9338493539482381, | |
| "eval_loss": 0.06378066539764404, | |
| "eval_runtime": 8.4193, | |
| "eval_samples_per_second": 71.74, | |
| "eval_steps_per_second": 23.993, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 3.960674157303371e-05, | |
| "loss": 0.046, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "learning_rate": 3.9325842696629214e-05, | |
| "loss": 0.0456, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 3.904494382022472e-05, | |
| "loss": 0.0448, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 3.876404494382023e-05, | |
| "loss": 0.0447, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 3.8483146067415735e-05, | |
| "loss": 0.0441, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "learning_rate": 3.8202247191011236e-05, | |
| "loss": 0.0438, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "eval_accuracy": 0.933936833374373, | |
| "eval_loss": 0.06545107811689377, | |
| "eval_runtime": 8.4022, | |
| "eval_samples_per_second": 71.886, | |
| "eval_steps_per_second": 24.041, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 3.792134831460674e-05, | |
| "loss": 0.0431, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "learning_rate": 3.764044943820225e-05, | |
| "loss": 0.0422, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 3.735955056179776e-05, | |
| "loss": 0.0422, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 5.13, | |
| "learning_rate": 3.7078651685393264e-05, | |
| "loss": 0.0408, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 3.6797752808988765e-05, | |
| "loss": 0.0402, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "learning_rate": 3.651685393258427e-05, | |
| "loss": 0.0405, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "eval_accuracy": 0.9344649499099286, | |
| "eval_loss": 0.06603477150201797, | |
| "eval_runtime": 8.3885, | |
| "eval_samples_per_second": 72.003, | |
| "eval_steps_per_second": 24.081, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 5.47, | |
| "learning_rate": 3.623595505617978e-05, | |
| "loss": 0.039, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 5.58, | |
| "learning_rate": 3.595505617977528e-05, | |
| "loss": 0.0387, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "learning_rate": 3.5674157303370787e-05, | |
| "loss": 0.0386, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "learning_rate": 3.5393258426966294e-05, | |
| "loss": 0.0385, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 5.91, | |
| "learning_rate": 3.51123595505618e-05, | |
| "loss": 0.0376, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 3.483146067415731e-05, | |
| "loss": 0.0378, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "eval_accuracy": 0.9341765918015578, | |
| "eval_loss": 0.06659498065710068, | |
| "eval_runtime": 8.3709, | |
| "eval_samples_per_second": 72.154, | |
| "eval_steps_per_second": 24.131, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 6.14, | |
| "learning_rate": 3.455056179775281e-05, | |
| "loss": 0.0366, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "learning_rate": 3.4269662921348316e-05, | |
| "loss": 0.0356, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 6.36, | |
| "learning_rate": 3.398876404494382e-05, | |
| "loss": 0.0351, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 6.47, | |
| "learning_rate": 3.370786516853933e-05, | |
| "loss": 0.0353, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 6.58, | |
| "learning_rate": 3.342696629213483e-05, | |
| "loss": 0.0348, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 6.69, | |
| "learning_rate": 3.314606741573034e-05, | |
| "loss": 0.0344, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.69, | |
| "eval_accuracy": 0.93427379116393, | |
| "eval_loss": 0.06691180914640427, | |
| "eval_runtime": 8.391, | |
| "eval_samples_per_second": 71.982, | |
| "eval_steps_per_second": 24.073, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.81, | |
| "learning_rate": 3.2865168539325845e-05, | |
| "loss": 0.0334, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.92, | |
| "learning_rate": 3.258426966292135e-05, | |
| "loss": 0.0342, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 3.230337078651686e-05, | |
| "loss": 0.0331, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "learning_rate": 3.202247191011236e-05, | |
| "loss": 0.0324, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "learning_rate": 3.1741573033707866e-05, | |
| "loss": 0.0314, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 7.36, | |
| "learning_rate": 3.1460674157303374e-05, | |
| "loss": 0.0323, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 7.36, | |
| "eval_accuracy": 0.9343839504412851, | |
| "eval_loss": 0.06778556853532791, | |
| "eval_runtime": 8.8059, | |
| "eval_samples_per_second": 68.591, | |
| "eval_steps_per_second": 22.939, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "learning_rate": 3.1179775280898874e-05, | |
| "loss": 0.0317, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "learning_rate": 3.089887640449438e-05, | |
| "loss": 0.0324, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 7.7, | |
| "learning_rate": 3.061797752808989e-05, | |
| "loss": 0.0318, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 7.81, | |
| "learning_rate": 3.0337078651685396e-05, | |
| "loss": 0.0313, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.92, | |
| "learning_rate": 3.0056179775280903e-05, | |
| "loss": 0.031, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "learning_rate": 2.9775280898876406e-05, | |
| "loss": 0.0307, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "eval_accuracy": 0.9343191508663703, | |
| "eval_loss": 0.06937924772500992, | |
| "eval_runtime": 8.3943, | |
| "eval_samples_per_second": 71.954, | |
| "eval_steps_per_second": 24.064, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 8.15, | |
| "learning_rate": 2.949438202247191e-05, | |
| "loss": 0.0304, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 8.26, | |
| "learning_rate": 2.9213483146067417e-05, | |
| "loss": 0.0304, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 8.37, | |
| "learning_rate": 2.893258426966292e-05, | |
| "loss": 0.0296, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "learning_rate": 2.8651685393258425e-05, | |
| "loss": 0.0295, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 8.59, | |
| "learning_rate": 2.8370786516853936e-05, | |
| "loss": 0.0299, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "learning_rate": 2.8089887640449443e-05, | |
| "loss": 0.0294, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "eval_accuracy": 0.9344681898886743, | |
| "eval_loss": 0.07063037902116776, | |
| "eval_runtime": 16.4408, | |
| "eval_samples_per_second": 36.738, | |
| "eval_steps_per_second": 12.286, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 8.81, | |
| "learning_rate": 2.7808988764044946e-05, | |
| "loss": 0.03, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 8.93, | |
| "learning_rate": 2.752808988764045e-05, | |
| "loss": 0.0296, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 2.7247191011235957e-05, | |
| "loss": 0.0288, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 9.15, | |
| "learning_rate": 2.696629213483146e-05, | |
| "loss": 0.0281, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 9.26, | |
| "learning_rate": 2.6685393258426965e-05, | |
| "loss": 0.0286, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 9.37, | |
| "learning_rate": 2.6404494382022472e-05, | |
| "loss": 0.0286, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 9.37, | |
| "eval_accuracy": 0.9342381513977268, | |
| "eval_loss": 0.0725179985165596, | |
| "eval_runtime": 16.4083, | |
| "eval_samples_per_second": 36.811, | |
| "eval_steps_per_second": 12.311, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 9.48, | |
| "learning_rate": 2.6123595505617983e-05, | |
| "loss": 0.0288, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "learning_rate": 2.5842696629213486e-05, | |
| "loss": 0.0284, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 9.71, | |
| "learning_rate": 2.556179775280899e-05, | |
| "loss": 0.0283, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 9.82, | |
| "learning_rate": 2.5280898876404497e-05, | |
| "loss": 0.0282, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 9.93, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0278, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 10.04, | |
| "learning_rate": 2.4719101123595505e-05, | |
| "loss": 0.0275, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 10.04, | |
| "eval_accuracy": 0.9343159108876246, | |
| "eval_loss": 0.0727309137582779, | |
| "eval_runtime": 16.3442, | |
| "eval_samples_per_second": 36.955, | |
| "eval_steps_per_second": 12.359, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 10.15, | |
| "learning_rate": 2.4438202247191012e-05, | |
| "loss": 0.0272, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 10.26, | |
| "learning_rate": 2.415730337078652e-05, | |
| "loss": 0.0271, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 10.38, | |
| "learning_rate": 2.3876404494382023e-05, | |
| "loss": 0.027, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 10.49, | |
| "learning_rate": 2.359550561797753e-05, | |
| "loss": 0.0275, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 10.6, | |
| "learning_rate": 2.3314606741573034e-05, | |
| "loss": 0.0271, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 10.71, | |
| "learning_rate": 2.303370786516854e-05, | |
| "loss": 0.0282, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 10.71, | |
| "eval_accuracy": 0.934247871333964, | |
| "eval_loss": 0.07324212789535522, | |
| "eval_runtime": 16.4154, | |
| "eval_samples_per_second": 36.795, | |
| "eval_steps_per_second": 12.306, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 10.82, | |
| "learning_rate": 2.2752808988764045e-05, | |
| "loss": 0.0263, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 10.93, | |
| "learning_rate": 2.2471910112359552e-05, | |
| "loss": 0.0266, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 2.2191011235955056e-05, | |
| "loss": 0.0269, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 11.16, | |
| "learning_rate": 2.1910112359550563e-05, | |
| "loss": 0.0266, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 11.27, | |
| "learning_rate": 2.1629213483146067e-05, | |
| "loss": 0.0259, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 11.38, | |
| "learning_rate": 2.1348314606741574e-05, | |
| "loss": 0.0264, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 11.38, | |
| "eval_accuracy": 0.9342802711214214, | |
| "eval_loss": 0.07354074716567993, | |
| "eval_runtime": 16.5992, | |
| "eval_samples_per_second": 36.387, | |
| "eval_steps_per_second": 12.169, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 11.49, | |
| "learning_rate": 2.1067415730337078e-05, | |
| "loss": 0.0265, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 11.6, | |
| "learning_rate": 2.0786516853932585e-05, | |
| "loss": 0.0262, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "learning_rate": 2.0505617977528092e-05, | |
| "loss": 0.0259, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 11.83, | |
| "learning_rate": 2.0224719101123596e-05, | |
| "loss": 0.026, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 11.94, | |
| "learning_rate": 1.99438202247191e-05, | |
| "loss": 0.026, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 1.9662921348314607e-05, | |
| "loss": 0.026, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "eval_accuracy": 0.9342413913764726, | |
| "eval_loss": 0.07500974833965302, | |
| "eval_runtime": 16.431, | |
| "eval_samples_per_second": 36.76, | |
| "eval_steps_per_second": 12.294, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 12.16, | |
| "learning_rate": 1.9382022471910114e-05, | |
| "loss": 0.0254, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 12.27, | |
| "learning_rate": 1.9101123595505618e-05, | |
| "loss": 0.0253, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 12.38, | |
| "learning_rate": 1.8820224719101125e-05, | |
| "loss": 0.0259, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "learning_rate": 1.8539325842696632e-05, | |
| "loss": 0.0253, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 12.61, | |
| "learning_rate": 1.8258426966292136e-05, | |
| "loss": 0.0253, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 12.72, | |
| "learning_rate": 1.797752808988764e-05, | |
| "loss": 0.0254, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 12.72, | |
| "eval_accuracy": 0.9343094309301331, | |
| "eval_loss": 0.07532921433448792, | |
| "eval_runtime": 16.5723, | |
| "eval_samples_per_second": 36.446, | |
| "eval_steps_per_second": 12.189, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 12.83, | |
| "learning_rate": 1.7696629213483147e-05, | |
| "loss": 0.0254, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 12.94, | |
| "learning_rate": 1.7415730337078654e-05, | |
| "loss": 0.0255, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 13.05, | |
| "learning_rate": 1.7134831460674158e-05, | |
| "loss": 0.0249, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 13.17, | |
| "learning_rate": 1.6853932584269665e-05, | |
| "loss": 0.0246, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 13.28, | |
| "learning_rate": 1.657303370786517e-05, | |
| "loss": 0.0245, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 13.39, | |
| "learning_rate": 1.6292134831460676e-05, | |
| "loss": 0.0244, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 13.39, | |
| "eval_accuracy": 0.9343515506538277, | |
| "eval_loss": 0.07460575550794601, | |
| "eval_runtime": 16.4167, | |
| "eval_samples_per_second": 36.792, | |
| "eval_steps_per_second": 12.305, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 13.5, | |
| "learning_rate": 1.601123595505618e-05, | |
| "loss": 0.025, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 13.61, | |
| "learning_rate": 1.5730337078651687e-05, | |
| "loss": 0.025, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 13.72, | |
| "learning_rate": 1.544943820224719e-05, | |
| "loss": 0.0244, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 13.84, | |
| "learning_rate": 1.5168539325842698e-05, | |
| "loss": 0.0247, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 13.95, | |
| "learning_rate": 1.4887640449438203e-05, | |
| "loss": 0.0248, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 14.06, | |
| "learning_rate": 1.4606741573033709e-05, | |
| "loss": 0.0242, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 14.06, | |
| "eval_accuracy": 0.9342964710151501, | |
| "eval_loss": 0.0752115547657013, | |
| "eval_runtime": 16.4268, | |
| "eval_samples_per_second": 36.769, | |
| "eval_steps_per_second": 12.297, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 14.17, | |
| "learning_rate": 1.4325842696629212e-05, | |
| "loss": 0.0243, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 14.28, | |
| "learning_rate": 1.4044943820224721e-05, | |
| "loss": 0.0243, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 14.39, | |
| "learning_rate": 1.3764044943820225e-05, | |
| "loss": 0.0245, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 14.5, | |
| "learning_rate": 1.348314606741573e-05, | |
| "loss": 0.0241, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 14.62, | |
| "learning_rate": 1.3202247191011236e-05, | |
| "loss": 0.0245, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 14.73, | |
| "learning_rate": 1.2921348314606743e-05, | |
| "loss": 0.024, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 14.73, | |
| "eval_accuracy": 0.9341701118440663, | |
| "eval_loss": 0.07579129934310913, | |
| "eval_runtime": 16.4351, | |
| "eval_samples_per_second": 36.751, | |
| "eval_steps_per_second": 12.291, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 14.84, | |
| "learning_rate": 1.2640449438202249e-05, | |
| "loss": 0.0238, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 14.95, | |
| "learning_rate": 1.2359550561797752e-05, | |
| "loss": 0.0239, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 15.06, | |
| "learning_rate": 1.207865168539326e-05, | |
| "loss": 0.0233, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 15.17, | |
| "learning_rate": 1.1797752808988765e-05, | |
| "loss": 0.0237, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 15.29, | |
| "learning_rate": 1.151685393258427e-05, | |
| "loss": 0.0238, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 15.4, | |
| "learning_rate": 1.1235955056179776e-05, | |
| "loss": 0.0239, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 15.4, | |
| "eval_accuracy": 0.934322390845116, | |
| "eval_loss": 0.07643470913171768, | |
| "eval_runtime": 16.4296, | |
| "eval_samples_per_second": 36.763, | |
| "eval_steps_per_second": 12.295, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 15.51, | |
| "learning_rate": 1.0955056179775282e-05, | |
| "loss": 0.0234, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 15.62, | |
| "learning_rate": 1.0674157303370787e-05, | |
| "loss": 0.0235, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 15.73, | |
| "learning_rate": 1.0393258426966292e-05, | |
| "loss": 0.0233, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 15.84, | |
| "learning_rate": 1.0112359550561798e-05, | |
| "loss": 0.0236, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 15.96, | |
| "learning_rate": 9.831460674157303e-06, | |
| "loss": 0.0236, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 16.07, | |
| "learning_rate": 9.550561797752809e-06, | |
| "loss": 0.0234, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 16.07, | |
| "eval_accuracy": 0.9342835111001672, | |
| "eval_loss": 0.07627058029174805, | |
| "eval_runtime": 16.539, | |
| "eval_samples_per_second": 36.52, | |
| "eval_steps_per_second": 12.214, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 16.18, | |
| "learning_rate": 9.269662921348316e-06, | |
| "loss": 0.0235, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 16.29, | |
| "learning_rate": 8.98876404494382e-06, | |
| "loss": 0.023, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 16.4, | |
| "learning_rate": 8.707865168539327e-06, | |
| "loss": 0.0236, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 16.51, | |
| "learning_rate": 8.426966292134832e-06, | |
| "loss": 0.0231, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 16.62, | |
| "learning_rate": 8.146067415730338e-06, | |
| "loss": 0.0234, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 16.74, | |
| "learning_rate": 7.865168539325843e-06, | |
| "loss": 0.0231, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 16.74, | |
| "eval_accuracy": 0.9343126709088788, | |
| "eval_loss": 0.07644984126091003, | |
| "eval_runtime": 16.3314, | |
| "eval_samples_per_second": 36.984, | |
| "eval_steps_per_second": 12.369, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 16.85, | |
| "learning_rate": 7.584269662921349e-06, | |
| "loss": 0.0229, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 16.96, | |
| "learning_rate": 7.303370786516854e-06, | |
| "loss": 0.0233, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 17.07, | |
| "learning_rate": 7.022471910112361e-06, | |
| "loss": 0.0231, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 17.18, | |
| "learning_rate": 6.741573033707865e-06, | |
| "loss": 0.023, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 17.29, | |
| "learning_rate": 6.460674157303372e-06, | |
| "loss": 0.0226, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 17.41, | |
| "learning_rate": 6.179775280898876e-06, | |
| "loss": 0.0226, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 17.41, | |
| "eval_accuracy": 0.9342932310364044, | |
| "eval_loss": 0.07702977955341339, | |
| "eval_runtime": 16.4049, | |
| "eval_samples_per_second": 36.818, | |
| "eval_steps_per_second": 12.313, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 17.52, | |
| "learning_rate": 5.8988764044943826e-06, | |
| "loss": 0.023, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 17.63, | |
| "learning_rate": 5.617977528089888e-06, | |
| "loss": 0.0227, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 17.74, | |
| "learning_rate": 5.3370786516853935e-06, | |
| "loss": 0.0233, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 17.85, | |
| "learning_rate": 5.056179775280899e-06, | |
| "loss": 0.0229, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 17.96, | |
| "learning_rate": 4.7752808988764044e-06, | |
| "loss": 0.0228, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 18.08, | |
| "learning_rate": 4.49438202247191e-06, | |
| "loss": 0.023, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 18.08, | |
| "eval_accuracy": 0.9343126709088788, | |
| "eval_loss": 0.07704292237758636, | |
| "eval_runtime": 16.4328, | |
| "eval_samples_per_second": 36.756, | |
| "eval_steps_per_second": 12.292, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 18.19, | |
| "learning_rate": 4.213483146067416e-06, | |
| "loss": 0.0223, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 18.3, | |
| "learning_rate": 3.932584269662922e-06, | |
| "loss": 0.0223, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 18.41, | |
| "learning_rate": 3.651685393258427e-06, | |
| "loss": 0.0229, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 18.52, | |
| "learning_rate": 3.3707865168539327e-06, | |
| "loss": 0.0225, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 18.63, | |
| "learning_rate": 3.089887640449438e-06, | |
| "loss": 0.0229, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 18.74, | |
| "learning_rate": 2.808988764044944e-06, | |
| "loss": 0.0227, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 18.74, | |
| "eval_accuracy": 0.9343094309301331, | |
| "eval_loss": 0.07714974880218506, | |
| "eval_runtime": 16.563, | |
| "eval_samples_per_second": 36.467, | |
| "eval_steps_per_second": 12.196, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 18.86, | |
| "learning_rate": 2.5280898876404495e-06, | |
| "loss": 0.0232, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 18.97, | |
| "learning_rate": 2.247191011235955e-06, | |
| "loss": 0.0224, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 19.08, | |
| "learning_rate": 1.966292134831461e-06, | |
| "loss": 0.0222, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 19.19, | |
| "learning_rate": 1.6853932584269663e-06, | |
| "loss": 0.0225, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 19.3, | |
| "learning_rate": 1.404494382022472e-06, | |
| "loss": 0.0223, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 19.41, | |
| "learning_rate": 1.1235955056179775e-06, | |
| "loss": 0.0221, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 19.41, | |
| "eval_accuracy": 0.934322390845116, | |
| "eval_loss": 0.07709047198295593, | |
| "eval_runtime": 16.3614, | |
| "eval_samples_per_second": 36.916, | |
| "eval_steps_per_second": 12.346, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 19.53, | |
| "learning_rate": 8.426966292134832e-07, | |
| "loss": 0.0231, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 19.64, | |
| "learning_rate": 5.617977528089887e-07, | |
| "loss": 0.0227, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 19.75, | |
| "learning_rate": 2.8089887640449437e-07, | |
| "loss": 0.0226, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 19.86, | |
| "learning_rate": 0.0, | |
| "loss": 0.0227, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 19.86, | |
| "step": 1780, | |
| "total_flos": 6.551573269655685e+17, | |
| "train_loss": 0.09398916166987312, | |
| "train_runtime": 60723.6526, | |
| "train_samples_per_second": 17.711, | |
| "train_steps_per_second": 0.029 | |
| } | |
| ], | |
| "max_steps": 1780, | |
| "num_train_epochs": 20, | |
| "total_flos": 6.551573269655685e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |