|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 1.9083181542197936, |
|
"global_step": 4000, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.0, |
|
"learning_rate": 1.8e-07, |
|
"loss": 0.692, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"learning_rate": 3.8e-07, |
|
"loss": 0.6931, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"learning_rate": 5.800000000000001e-07, |
|
"loss": 0.6938, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"learning_rate": 7.8e-07, |
|
"loss": 0.6928, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"learning_rate": 9.800000000000001e-07, |
|
"loss": 0.6936, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"learning_rate": 1.1800000000000001e-06, |
|
"loss": 0.6944, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"learning_rate": 1.3800000000000001e-06, |
|
"loss": 0.6936, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"learning_rate": 1.5800000000000001e-06, |
|
"loss": 0.6926, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.04, |
|
"learning_rate": 1.7800000000000001e-06, |
|
"loss": 0.6921, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 0.05, |
|
"learning_rate": 1.98e-06, |
|
"loss": 0.6921, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.05, |
|
"learning_rate": 2.1800000000000003e-06, |
|
"loss": 0.6924, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"learning_rate": 2.38e-06, |
|
"loss": 0.6924, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"learning_rate": 2.5800000000000003e-06, |
|
"loss": 0.6902, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 0.07, |
|
"learning_rate": 2.7800000000000005e-06, |
|
"loss": 0.6914, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 0.07, |
|
"learning_rate": 2.9800000000000003e-06, |
|
"loss": 0.6907, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.08, |
|
"learning_rate": 3.1800000000000005e-06, |
|
"loss": 0.6893, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 0.08, |
|
"learning_rate": 3.3800000000000007e-06, |
|
"loss": 0.6867, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 0.09, |
|
"learning_rate": 3.58e-06, |
|
"loss": 0.683, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 0.09, |
|
"learning_rate": 3.7800000000000002e-06, |
|
"loss": 0.677, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"learning_rate": 3.980000000000001e-06, |
|
"loss": 0.6627, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"eval_webgpt_accuracy": 0.5160878447395302, |
|
"eval_webgpt_loss": 0.69077467918396, |
|
"eval_webgpt_runtime": 62.8216, |
|
"eval_webgpt_samples_per_second": 62.335, |
|
"eval_webgpt_steps_per_second": 6.24, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"eval_hfsummary_accuracy": 0.6221624399238279, |
|
"eval_hfsummary_loss": 0.6861704587936401, |
|
"eval_hfsummary_runtime": 1239.8418, |
|
"eval_hfsummary_samples_per_second": 26.683, |
|
"eval_hfsummary_steps_per_second": 2.669, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"eval_gptsynthetic_accuracy": 0.9957767722473605, |
|
"eval_gptsynthetic_loss": 0.3938244879245758, |
|
"eval_gptsynthetic_runtime": 45.6545, |
|
"eval_gptsynthetic_samples_per_second": 72.611, |
|
"eval_gptsynthetic_steps_per_second": 7.272, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"learning_rate": 4.18e-06, |
|
"loss": 0.6481, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 0.1, |
|
"learning_rate": 4.38e-06, |
|
"loss": 0.624, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 0.11, |
|
"learning_rate": 4.58e-06, |
|
"loss": 0.6219, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 0.11, |
|
"learning_rate": 4.78e-06, |
|
"loss": 0.6079, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"learning_rate": 4.980000000000001e-06, |
|
"loss": 0.5815, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"learning_rate": 5.18e-06, |
|
"loss": 0.5714, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 0.13, |
|
"learning_rate": 5.380000000000001e-06, |
|
"loss": 0.5904, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 0.13, |
|
"learning_rate": 5.580000000000001e-06, |
|
"loss": 0.5796, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"learning_rate": 5.78e-06, |
|
"loss": 0.5649, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"learning_rate": 5.98e-06, |
|
"loss": 0.581, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"learning_rate": 6.18e-06, |
|
"loss": 0.5481, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"learning_rate": 6.380000000000001e-06, |
|
"loss": 0.5839, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 0.16, |
|
"learning_rate": 6.5800000000000005e-06, |
|
"loss": 0.5552, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 0.16, |
|
"learning_rate": 6.780000000000001e-06, |
|
"loss": 0.5654, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 0.17, |
|
"learning_rate": 6.98e-06, |
|
"loss": 0.5618, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 0.17, |
|
"learning_rate": 7.180000000000001e-06, |
|
"loss": 0.5528, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 0.18, |
|
"learning_rate": 7.3800000000000005e-06, |
|
"loss": 0.5341, |
|
"step": 370 |
|
}, |
|
{ |
|
"epoch": 0.18, |
|
"learning_rate": 7.58e-06, |
|
"loss": 0.5458, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 0.19, |
|
"learning_rate": 7.78e-06, |
|
"loss": 0.5335, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 0.19, |
|
"learning_rate": 7.980000000000002e-06, |
|
"loss": 0.5285, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.19, |
|
"eval_webgpt_accuracy": 0.5617977528089888, |
|
"eval_webgpt_loss": 0.7310989499092102, |
|
"eval_webgpt_runtime": 63.0205, |
|
"eval_webgpt_samples_per_second": 62.139, |
|
"eval_webgpt_steps_per_second": 6.22, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.19, |
|
"eval_hfsummary_accuracy": 0.6381525254662516, |
|
"eval_hfsummary_loss": 0.6372680068016052, |
|
"eval_hfsummary_runtime": 1238.3216, |
|
"eval_hfsummary_samples_per_second": 26.716, |
|
"eval_hfsummary_steps_per_second": 2.672, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.19, |
|
"eval_gptsynthetic_accuracy": 0.9981900452488688, |
|
"eval_gptsynthetic_loss": 0.026522908359766006, |
|
"eval_gptsynthetic_runtime": 45.6333, |
|
"eval_gptsynthetic_samples_per_second": 72.644, |
|
"eval_gptsynthetic_steps_per_second": 7.275, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"learning_rate": 8.18e-06, |
|
"loss": 0.5663, |
|
"step": 410 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"learning_rate": 8.380000000000001e-06, |
|
"loss": 0.55, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"learning_rate": 8.580000000000001e-06, |
|
"loss": 0.5465, |
|
"step": 430 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"learning_rate": 8.78e-06, |
|
"loss": 0.5409, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"learning_rate": 8.98e-06, |
|
"loss": 0.5496, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"learning_rate": 9.180000000000002e-06, |
|
"loss": 0.5334, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"learning_rate": 9.38e-06, |
|
"loss": 0.5108, |
|
"step": 470 |
|
}, |
|
{ |
|
"epoch": 0.23, |
|
"learning_rate": 9.58e-06, |
|
"loss": 0.5306, |
|
"step": 480 |
|
}, |
|
{ |
|
"epoch": 0.23, |
|
"learning_rate": 9.780000000000001e-06, |
|
"loss": 0.5379, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 0.24, |
|
"learning_rate": 9.980000000000001e-06, |
|
"loss": 0.5288, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.24, |
|
"learning_rate": 9.999967852222371e-06, |
|
"loss": 0.5362, |
|
"step": 510 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"learning_rate": 9.999856724632902e-06, |
|
"loss": 0.5245, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"learning_rate": 9.99966622210923e-06, |
|
"loss": 0.5424, |
|
"step": 530 |
|
}, |
|
{ |
|
"epoch": 0.26, |
|
"learning_rate": 9.999396347675668e-06, |
|
"loss": 0.5161, |
|
"step": 540 |
|
}, |
|
{ |
|
"epoch": 0.26, |
|
"learning_rate": 9.99904710561659e-06, |
|
"loss": 0.5295, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"learning_rate": 9.998618501476367e-06, |
|
"loss": 0.5049, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"learning_rate": 9.998110542059283e-06, |
|
"loss": 0.5136, |
|
"step": 570 |
|
}, |
|
{ |
|
"epoch": 0.28, |
|
"learning_rate": 9.997523235429414e-06, |
|
"loss": 0.5059, |
|
"step": 580 |
|
}, |
|
{ |
|
"epoch": 0.28, |
|
"learning_rate": 9.996856590910513e-06, |
|
"loss": 0.5268, |
|
"step": 590 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"learning_rate": 9.996110619085859e-06, |
|
"loss": 0.515, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"eval_webgpt_accuracy": 0.5640960163432074, |
|
"eval_webgpt_loss": 0.7018715143203735, |
|
"eval_webgpt_runtime": 63.203, |
|
"eval_webgpt_samples_per_second": 61.959, |
|
"eval_webgpt_steps_per_second": 6.202, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"eval_hfsummary_accuracy": 0.6488831121724148, |
|
"eval_hfsummary_loss": 0.6238470077514648, |
|
"eval_hfsummary_runtime": 1239.6473, |
|
"eval_hfsummary_samples_per_second": 26.687, |
|
"eval_hfsummary_steps_per_second": 2.669, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"eval_gptsynthetic_accuracy": 0.9978883861236802, |
|
"eval_gptsynthetic_loss": 0.031191527843475342, |
|
"eval_gptsynthetic_runtime": 45.7943, |
|
"eval_gptsynthetic_samples_per_second": 72.389, |
|
"eval_gptsynthetic_steps_per_second": 7.25, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"learning_rate": 9.995285331798082e-06, |
|
"loss": 0.4913, |
|
"step": 610 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"learning_rate": 9.994380742148981e-06, |
|
"loss": 0.4957, |
|
"step": 620 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"learning_rate": 9.993396864499321e-06, |
|
"loss": 0.5041, |
|
"step": 630 |
|
}, |
|
{ |
|
"epoch": 0.31, |
|
"learning_rate": 9.992333714468589e-06, |
|
"loss": 0.5204, |
|
"step": 640 |
|
}, |
|
{ |
|
"epoch": 0.31, |
|
"learning_rate": 9.991191308934765e-06, |
|
"loss": 0.4964, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 0.31, |
|
"learning_rate": 9.989969666034039e-06, |
|
"loss": 0.5054, |
|
"step": 660 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"learning_rate": 9.988668805160533e-06, |
|
"loss": 0.5156, |
|
"step": 670 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"learning_rate": 9.987288746965987e-06, |
|
"loss": 0.4875, |
|
"step": 680 |
|
}, |
|
{ |
|
"epoch": 0.33, |
|
"learning_rate": 9.985829513359437e-06, |
|
"loss": 0.5002, |
|
"step": 690 |
|
}, |
|
{ |
|
"epoch": 0.33, |
|
"learning_rate": 9.984291127506857e-06, |
|
"loss": 0.5103, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 0.34, |
|
"learning_rate": 9.982673613830807e-06, |
|
"loss": 0.488, |
|
"step": 710 |
|
}, |
|
{ |
|
"epoch": 0.34, |
|
"learning_rate": 9.980976998010028e-06, |
|
"loss": 0.5091, |
|
"step": 720 |
|
}, |
|
{ |
|
"epoch": 0.35, |
|
"learning_rate": 9.979201306979047e-06, |
|
"loss": 0.5008, |
|
"step": 730 |
|
}, |
|
{ |
|
"epoch": 0.35, |
|
"learning_rate": 9.977346568927742e-06, |
|
"loss": 0.4933, |
|
"step": 740 |
|
}, |
|
{ |
|
"epoch": 0.36, |
|
"learning_rate": 9.9754128133009e-06, |
|
"loss": 0.5208, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 0.36, |
|
"learning_rate": 9.973400070797745e-06, |
|
"loss": 0.5036, |
|
"step": 760 |
|
}, |
|
{ |
|
"epoch": 0.37, |
|
"learning_rate": 9.971308373371453e-06, |
|
"loss": 0.4935, |
|
"step": 770 |
|
}, |
|
{ |
|
"epoch": 0.37, |
|
"learning_rate": 9.969137754228648e-06, |
|
"loss": 0.5026, |
|
"step": 780 |
|
}, |
|
{ |
|
"epoch": 0.38, |
|
"learning_rate": 9.966888247828864e-06, |
|
"loss": 0.5047, |
|
"step": 790 |
|
}, |
|
{ |
|
"epoch": 0.38, |
|
"learning_rate": 9.964559889884011e-06, |
|
"loss": 0.4785, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 0.38, |
|
"eval_webgpt_accuracy": 0.5684371807967313, |
|
"eval_webgpt_loss": 0.6995264887809753, |
|
"eval_webgpt_runtime": 63.093, |
|
"eval_webgpt_samples_per_second": 62.067, |
|
"eval_webgpt_steps_per_second": 6.213, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 0.38, |
|
"eval_hfsummary_accuracy": 0.6444397424659191, |
|
"eval_hfsummary_loss": 0.6311432719230652, |
|
"eval_hfsummary_runtime": 1241.3962, |
|
"eval_hfsummary_samples_per_second": 26.65, |
|
"eval_hfsummary_steps_per_second": 2.666, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 0.38, |
|
"eval_gptsynthetic_accuracy": 0.9978883861236802, |
|
"eval_gptsynthetic_loss": 0.016028160229325294, |
|
"eval_gptsynthetic_runtime": 45.7378, |
|
"eval_gptsynthetic_samples_per_second": 72.478, |
|
"eval_gptsynthetic_steps_per_second": 7.259, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 0.39, |
|
"learning_rate": 9.962152717357796e-06, |
|
"loss": 0.4752, |
|
"step": 810 |
|
}, |
|
{ |
|
"epoch": 0.39, |
|
"learning_rate": 9.959666768465153e-06, |
|
"loss": 0.4657, |
|
"step": 820 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"learning_rate": 9.957102082671612e-06, |
|
"loss": 0.4935, |
|
"step": 830 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"learning_rate": 9.954458700692699e-06, |
|
"loss": 0.4821, |
|
"step": 840 |
|
}, |
|
{ |
|
"epoch": 0.41, |
|
"learning_rate": 9.951736664493266e-06, |
|
"loss": 0.4949, |
|
"step": 850 |
|
}, |
|
{ |
|
"epoch": 0.41, |
|
"learning_rate": 9.948936017286844e-06, |
|
"loss": 0.4894, |
|
"step": 860 |
|
}, |
|
{ |
|
"epoch": 0.42, |
|
"learning_rate": 9.94605680353494e-06, |
|
"loss": 0.5032, |
|
"step": 870 |
|
}, |
|
{ |
|
"epoch": 0.42, |
|
"learning_rate": 9.943099068946348e-06, |
|
"loss": 0.4724, |
|
"step": 880 |
|
}, |
|
{ |
|
"epoch": 0.42, |
|
"learning_rate": 9.940062860476406e-06, |
|
"loss": 0.4871, |
|
"step": 890 |
|
}, |
|
{ |
|
"epoch": 0.43, |
|
"learning_rate": 9.936948226326273e-06, |
|
"loss": 0.4919, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 0.43, |
|
"learning_rate": 9.933755215942137e-06, |
|
"loss": 0.4763, |
|
"step": 910 |
|
}, |
|
{ |
|
"epoch": 0.44, |
|
"learning_rate": 9.930483880014448e-06, |
|
"loss": 0.4928, |
|
"step": 920 |
|
}, |
|
{ |
|
"epoch": 0.44, |
|
"learning_rate": 9.92713427047711e-06, |
|
"loss": 0.4608, |
|
"step": 930 |
|
}, |
|
{ |
|
"epoch": 0.45, |
|
"learning_rate": 9.923706440506654e-06, |
|
"loss": 0.491, |
|
"step": 940 |
|
}, |
|
{ |
|
"epoch": 0.45, |
|
"learning_rate": 9.920200444521395e-06, |
|
"loss": 0.4624, |
|
"step": 950 |
|
}, |
|
{ |
|
"epoch": 0.46, |
|
"learning_rate": 9.916616338180569e-06, |
|
"loss": 0.5028, |
|
"step": 960 |
|
}, |
|
{ |
|
"epoch": 0.46, |
|
"learning_rate": 9.912954178383445e-06, |
|
"loss": 0.4814, |
|
"step": 970 |
|
}, |
|
{ |
|
"epoch": 0.47, |
|
"learning_rate": 9.90921402326843e-06, |
|
"loss": 0.4901, |
|
"step": 980 |
|
}, |
|
{ |
|
"epoch": 0.47, |
|
"learning_rate": 9.905395932212139e-06, |
|
"loss": 0.459, |
|
"step": 990 |
|
}, |
|
{ |
|
"epoch": 0.48, |
|
"learning_rate": 9.901499965828451e-06, |
|
"loss": 0.4624, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.48, |
|
"eval_webgpt_accuracy": 0.5661389172625128, |
|
"eval_webgpt_loss": 0.6934247016906738, |
|
"eval_webgpt_runtime": 63.1076, |
|
"eval_webgpt_samples_per_second": 62.053, |
|
"eval_webgpt_steps_per_second": 6.212, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.48, |
|
"eval_hfsummary_accuracy": 0.6628177613880241, |
|
"eval_hfsummary_loss": 0.6276853680610657, |
|
"eval_hfsummary_runtime": 1241.5998, |
|
"eval_hfsummary_samples_per_second": 26.645, |
|
"eval_hfsummary_steps_per_second": 2.665, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.48, |
|
"eval_gptsynthetic_accuracy": 0.9975867269984917, |
|
"eval_gptsynthetic_loss": 0.015120746567845345, |
|
"eval_gptsynthetic_runtime": 45.6555, |
|
"eval_gptsynthetic_samples_per_second": 72.609, |
|
"eval_gptsynthetic_steps_per_second": 7.272, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.48, |
|
"learning_rate": 9.897526185967557e-06, |
|
"loss": 0.4729, |
|
"step": 1010 |
|
}, |
|
{ |
|
"epoch": 0.49, |
|
"learning_rate": 9.893474655714962e-06, |
|
"loss": 0.4767, |
|
"step": 1020 |
|
}, |
|
{ |
|
"epoch": 0.49, |
|
"learning_rate": 9.889345439390508e-06, |
|
"loss": 0.4737, |
|
"step": 1030 |
|
}, |
|
{ |
|
"epoch": 0.5, |
|
"learning_rate": 9.88513860254732e-06, |
|
"loss": 0.4795, |
|
"step": 1040 |
|
}, |
|
{ |
|
"epoch": 0.5, |
|
"learning_rate": 9.880854211970794e-06, |
|
"loss": 0.4674, |
|
"step": 1050 |
|
}, |
|
{ |
|
"epoch": 0.51, |
|
"learning_rate": 9.876492335677523e-06, |
|
"loss": 0.4605, |
|
"step": 1060 |
|
}, |
|
{ |
|
"epoch": 0.51, |
|
"learning_rate": 9.872053042914223e-06, |
|
"loss": 0.4347, |
|
"step": 1070 |
|
}, |
|
{ |
|
"epoch": 0.52, |
|
"learning_rate": 9.867536404156621e-06, |
|
"loss": 0.4614, |
|
"step": 1080 |
|
}, |
|
{ |
|
"epoch": 0.52, |
|
"learning_rate": 9.862942491108358e-06, |
|
"loss": 0.4849, |
|
"step": 1090 |
|
}, |
|
{ |
|
"epoch": 0.52, |
|
"learning_rate": 9.858271376699832e-06, |
|
"loss": 0.4715, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 0.53, |
|
"learning_rate": 9.853523135087047e-06, |
|
"loss": 0.4768, |
|
"step": 1110 |
|
}, |
|
{ |
|
"epoch": 0.53, |
|
"learning_rate": 9.848697841650435e-06, |
|
"loss": 0.4686, |
|
"step": 1120 |
|
}, |
|
{ |
|
"epoch": 0.54, |
|
"learning_rate": 9.843795572993664e-06, |
|
"loss": 0.4521, |
|
"step": 1130 |
|
}, |
|
{ |
|
"epoch": 0.54, |
|
"learning_rate": 9.83881640694241e-06, |
|
"loss": 0.4765, |
|
"step": 1140 |
|
}, |
|
{ |
|
"epoch": 0.55, |
|
"learning_rate": 9.833760422543137e-06, |
|
"loss": 0.4829, |
|
"step": 1150 |
|
}, |
|
{ |
|
"epoch": 0.55, |
|
"learning_rate": 9.828627700061833e-06, |
|
"loss": 0.4587, |
|
"step": 1160 |
|
}, |
|
{ |
|
"epoch": 0.56, |
|
"learning_rate": 9.82341832098273e-06, |
|
"loss": 0.4684, |
|
"step": 1170 |
|
}, |
|
{ |
|
"epoch": 0.56, |
|
"learning_rate": 9.818132368007022e-06, |
|
"loss": 0.4443, |
|
"step": 1180 |
|
}, |
|
{ |
|
"epoch": 0.57, |
|
"learning_rate": 9.812769925051553e-06, |
|
"loss": 0.4898, |
|
"step": 1190 |
|
}, |
|
{ |
|
"epoch": 0.57, |
|
"learning_rate": 9.807331077247468e-06, |
|
"loss": 0.44, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 0.57, |
|
"eval_webgpt_accuracy": 0.5605209397344229, |
|
"eval_webgpt_loss": 0.6905099749565125, |
|
"eval_webgpt_runtime": 63.2036, |
|
"eval_webgpt_samples_per_second": 61.959, |
|
"eval_webgpt_steps_per_second": 6.202, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 0.57, |
|
"eval_hfsummary_accuracy": 0.664419792642747, |
|
"eval_hfsummary_loss": 0.6330422163009644, |
|
"eval_hfsummary_runtime": 1242.5416, |
|
"eval_hfsummary_samples_per_second": 26.625, |
|
"eval_hfsummary_steps_per_second": 2.663, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 0.57, |
|
"eval_gptsynthetic_accuracy": 0.9981900452488688, |
|
"eval_gptsynthetic_loss": 0.011154625564813614, |
|
"eval_gptsynthetic_runtime": 45.6845, |
|
"eval_gptsynthetic_samples_per_second": 72.563, |
|
"eval_gptsynthetic_steps_per_second": 7.267, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 0.58, |
|
"learning_rate": 9.80181591093888e-06, |
|
"loss": 0.4648, |
|
"step": 1210 |
|
}, |
|
{ |
|
"epoch": 0.58, |
|
"learning_rate": 9.796224513681495e-06, |
|
"loss": 0.4483, |
|
"step": 1220 |
|
}, |
|
{ |
|
"epoch": 0.59, |
|
"learning_rate": 9.79055697424121e-06, |
|
"loss": 0.4707, |
|
"step": 1230 |
|
}, |
|
{ |
|
"epoch": 0.59, |
|
"learning_rate": 9.784813382592716e-06, |
|
"loss": 0.4638, |
|
"step": 1240 |
|
}, |
|
{ |
|
"epoch": 0.6, |
|
"learning_rate": 9.77899382991807e-06, |
|
"loss": 0.4366, |
|
"step": 1250 |
|
}, |
|
{ |
|
"epoch": 0.6, |
|
"learning_rate": 9.773098408605242e-06, |
|
"loss": 0.4457, |
|
"step": 1260 |
|
}, |
|
{ |
|
"epoch": 0.61, |
|
"learning_rate": 9.76712721224665e-06, |
|
"loss": 0.4363, |
|
"step": 1270 |
|
}, |
|
{ |
|
"epoch": 0.61, |
|
"learning_rate": 9.761080335637672e-06, |
|
"loss": 0.4577, |
|
"step": 1280 |
|
}, |
|
{ |
|
"epoch": 0.62, |
|
"learning_rate": 9.754957874775146e-06, |
|
"loss": 0.4538, |
|
"step": 1290 |
|
}, |
|
{ |
|
"epoch": 0.62, |
|
"learning_rate": 9.748759926855845e-06, |
|
"loss": 0.4499, |
|
"step": 1300 |
|
}, |
|
{ |
|
"epoch": 0.62, |
|
"learning_rate": 9.742486590274927e-06, |
|
"loss": 0.4466, |
|
"step": 1310 |
|
}, |
|
{ |
|
"epoch": 0.63, |
|
"learning_rate": 9.736137964624381e-06, |
|
"loss": 0.4459, |
|
"step": 1320 |
|
}, |
|
{ |
|
"epoch": 0.63, |
|
"learning_rate": 9.729714150691448e-06, |
|
"loss": 0.4544, |
|
"step": 1330 |
|
}, |
|
{ |
|
"epoch": 0.64, |
|
"learning_rate": 9.723215250457009e-06, |
|
"loss": 0.4453, |
|
"step": 1340 |
|
}, |
|
{ |
|
"epoch": 0.64, |
|
"learning_rate": 9.716641367093976e-06, |
|
"loss": 0.4614, |
|
"step": 1350 |
|
}, |
|
{ |
|
"epoch": 0.65, |
|
"learning_rate": 9.709992604965652e-06, |
|
"loss": 0.4863, |
|
"step": 1360 |
|
}, |
|
{ |
|
"epoch": 0.65, |
|
"learning_rate": 9.703269069624075e-06, |
|
"loss": 0.4432, |
|
"step": 1370 |
|
}, |
|
{ |
|
"epoch": 0.66, |
|
"learning_rate": 9.696470867808333e-06, |
|
"loss": 0.438, |
|
"step": 1380 |
|
}, |
|
{ |
|
"epoch": 0.66, |
|
"learning_rate": 9.68959810744289e-06, |
|
"loss": 0.4593, |
|
"step": 1390 |
|
}, |
|
{ |
|
"epoch": 0.67, |
|
"learning_rate": 9.682650897635845e-06, |
|
"loss": 0.4463, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 0.67, |
|
"eval_webgpt_accuracy": 0.5617977528089888, |
|
"eval_webgpt_loss": 0.6875464916229248, |
|
"eval_webgpt_runtime": 63.2336, |
|
"eval_webgpt_samples_per_second": 61.929, |
|
"eval_webgpt_steps_per_second": 6.199, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 0.67, |
|
"eval_hfsummary_accuracy": 0.6741528881903093, |
|
"eval_hfsummary_loss": 0.6355233788490295, |
|
"eval_hfsummary_runtime": 1241.0977, |
|
"eval_hfsummary_samples_per_second": 26.656, |
|
"eval_hfsummary_steps_per_second": 2.666, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 0.67, |
|
"eval_gptsynthetic_accuracy": 0.9975867269984917, |
|
"eval_gptsynthetic_loss": 0.010252352803945541, |
|
"eval_gptsynthetic_runtime": 45.7898, |
|
"eval_gptsynthetic_samples_per_second": 72.396, |
|
"eval_gptsynthetic_steps_per_second": 7.251, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 0.67, |
|
"learning_rate": 9.675629348677232e-06, |
|
"loss": 0.4454, |
|
"step": 1410 |
|
}, |
|
{ |
|
"epoch": 0.68, |
|
"learning_rate": 9.668533572037236e-06, |
|
"loss": 0.4444, |
|
"step": 1420 |
|
}, |
|
{ |
|
"epoch": 0.68, |
|
"learning_rate": 9.661363680364445e-06, |
|
"loss": 0.4622, |
|
"step": 1430 |
|
}, |
|
{ |
|
"epoch": 0.69, |
|
"learning_rate": 9.65411978748406e-06, |
|
"loss": 0.4591, |
|
"step": 1440 |
|
}, |
|
{ |
|
"epoch": 0.69, |
|
"learning_rate": 9.64680200839608e-06, |
|
"loss": 0.4462, |
|
"step": 1450 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"learning_rate": 9.63941045927348e-06, |
|
"loss": 0.4423, |
|
"step": 1460 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"learning_rate": 9.631945257460369e-06, |
|
"loss": 0.4302, |
|
"step": 1470 |
|
}, |
|
{ |
|
"epoch": 0.71, |
|
"learning_rate": 9.62440652147012e-06, |
|
"loss": 0.4668, |
|
"step": 1480 |
|
}, |
|
{ |
|
"epoch": 0.71, |
|
"learning_rate": 9.616794370983498e-06, |
|
"loss": 0.4252, |
|
"step": 1490 |
|
}, |
|
{ |
|
"epoch": 0.72, |
|
"learning_rate": 9.609108926846759e-06, |
|
"loss": 0.4563, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 0.72, |
|
"learning_rate": 9.60135031106972e-06, |
|
"loss": 0.4396, |
|
"step": 1510 |
|
}, |
|
{ |
|
"epoch": 0.73, |
|
"learning_rate": 9.593518646823833e-06, |
|
"loss": 0.463, |
|
"step": 1520 |
|
}, |
|
{ |
|
"epoch": 0.73, |
|
"learning_rate": 9.58561405844023e-06, |
|
"loss": 0.4292, |
|
"step": 1530 |
|
}, |
|
{ |
|
"epoch": 0.73, |
|
"learning_rate": 9.577636671407743e-06, |
|
"loss": 0.4669, |
|
"step": 1540 |
|
}, |
|
{ |
|
"epoch": 0.74, |
|
"learning_rate": 9.569586612370913e-06, |
|
"loss": 0.4648, |
|
"step": 1550 |
|
}, |
|
{ |
|
"epoch": 0.74, |
|
"learning_rate": 9.561464009127988e-06, |
|
"loss": 0.4318, |
|
"step": 1560 |
|
}, |
|
{ |
|
"epoch": 0.75, |
|
"learning_rate": 9.553268990628877e-06, |
|
"loss": 0.4274, |
|
"step": 1570 |
|
}, |
|
{ |
|
"epoch": 0.75, |
|
"learning_rate": 9.545001686973123e-06, |
|
"loss": 0.4532, |
|
"step": 1580 |
|
}, |
|
{ |
|
"epoch": 0.76, |
|
"learning_rate": 9.53666222940782e-06, |
|
"loss": 0.4476, |
|
"step": 1590 |
|
}, |
|
{ |
|
"epoch": 0.76, |
|
"learning_rate": 9.528250750325545e-06, |
|
"loss": 0.4477, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 0.76, |
|
"eval_webgpt_accuracy": 0.5801838610827375, |
|
"eval_webgpt_loss": 0.6804662346839905, |
|
"eval_webgpt_runtime": 63.2449, |
|
"eval_webgpt_samples_per_second": 61.918, |
|
"eval_webgpt_steps_per_second": 6.198, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 0.76, |
|
"eval_hfsummary_accuracy": 0.6604298280083426, |
|
"eval_hfsummary_loss": 0.6250379085540771, |
|
"eval_hfsummary_runtime": 1240.5251, |
|
"eval_hfsummary_samples_per_second": 26.669, |
|
"eval_hfsummary_steps_per_second": 2.667, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 0.76, |
|
"eval_gptsynthetic_accuracy": 0.9975867269984917, |
|
"eval_gptsynthetic_loss": 0.01225399598479271, |
|
"eval_gptsynthetic_runtime": 45.6829, |
|
"eval_gptsynthetic_samples_per_second": 72.565, |
|
"eval_gptsynthetic_steps_per_second": 7.267, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 0.77, |
|
"learning_rate": 9.519767383262244e-06, |
|
"loss": 0.4265, |
|
"step": 1610 |
|
}, |
|
{ |
|
"epoch": 0.77, |
|
"learning_rate": 9.511212262895113e-06, |
|
"loss": 0.4509, |
|
"step": 1620 |
|
}, |
|
{ |
|
"epoch": 0.78, |
|
"learning_rate": 9.502585525040469e-06, |
|
"loss": 0.4159, |
|
"step": 1630 |
|
}, |
|
{ |
|
"epoch": 0.78, |
|
"learning_rate": 9.493887306651585e-06, |
|
"loss": 0.4241, |
|
"step": 1640 |
|
}, |
|
{ |
|
"epoch": 0.79, |
|
"learning_rate": 9.485117745816523e-06, |
|
"loss": 0.4186, |
|
"step": 1650 |
|
}, |
|
{ |
|
"epoch": 0.79, |
|
"learning_rate": 9.476276981755933e-06, |
|
"loss": 0.4334, |
|
"step": 1660 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"learning_rate": 9.467365154820853e-06, |
|
"loss": 0.4508, |
|
"step": 1670 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"learning_rate": 9.45838240649047e-06, |
|
"loss": 0.4227, |
|
"step": 1680 |
|
}, |
|
{ |
|
"epoch": 0.81, |
|
"learning_rate": 9.44932887936988e-06, |
|
"loss": 0.4391, |
|
"step": 1690 |
|
}, |
|
{ |
|
"epoch": 0.81, |
|
"learning_rate": 9.440204717187829e-06, |
|
"loss": 0.4461, |
|
"step": 1700 |
|
}, |
|
{ |
|
"epoch": 0.82, |
|
"learning_rate": 9.43101006479442e-06, |
|
"loss": 0.4334, |
|
"step": 1710 |
|
}, |
|
{ |
|
"epoch": 0.82, |
|
"learning_rate": 9.421745068158823e-06, |
|
"loss": 0.4095, |
|
"step": 1720 |
|
}, |
|
{ |
|
"epoch": 0.83, |
|
"learning_rate": 9.41240987436695e-06, |
|
"loss": 0.4629, |
|
"step": 1730 |
|
}, |
|
{ |
|
"epoch": 0.83, |
|
"learning_rate": 9.40300463161913e-06, |
|
"loss": 0.433, |
|
"step": 1740 |
|
}, |
|
{ |
|
"epoch": 0.83, |
|
"learning_rate": 9.393529489227745e-06, |
|
"loss": 0.4323, |
|
"step": 1750 |
|
}, |
|
{ |
|
"epoch": 0.84, |
|
"learning_rate": 9.383984597614867e-06, |
|
"loss": 0.4339, |
|
"step": 1760 |
|
}, |
|
{ |
|
"epoch": 0.84, |
|
"learning_rate": 9.374370108309862e-06, |
|
"loss": 0.4263, |
|
"step": 1770 |
|
}, |
|
{ |
|
"epoch": 0.85, |
|
"learning_rate": 9.364686173947e-06, |
|
"loss": 0.4095, |
|
"step": 1780 |
|
}, |
|
{ |
|
"epoch": 0.85, |
|
"learning_rate": 9.354932948263014e-06, |
|
"loss": 0.4284, |
|
"step": 1790 |
|
}, |
|
{ |
|
"epoch": 0.86, |
|
"learning_rate": 9.345110586094672e-06, |
|
"loss": 0.4398, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 0.86, |
|
"eval_webgpt_accuracy": 0.5873340143003064, |
|
"eval_webgpt_loss": 0.6802140474319458, |
|
"eval_webgpt_runtime": 63.2638, |
|
"eval_webgpt_samples_per_second": 61.9, |
|
"eval_webgpt_steps_per_second": 6.196, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 0.86, |
|
"eval_hfsummary_accuracy": 0.6609739140948523, |
|
"eval_hfsummary_loss": 0.6455074548721313, |
|
"eval_hfsummary_runtime": 1241.9993, |
|
"eval_hfsummary_samples_per_second": 26.637, |
|
"eval_hfsummary_steps_per_second": 2.664, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 0.86, |
|
"eval_gptsynthetic_accuracy": 0.9975867269984917, |
|
"eval_gptsynthetic_loss": 0.007792285177856684, |
|
"eval_gptsynthetic_runtime": 45.7802, |
|
"eval_gptsynthetic_samples_per_second": 72.411, |
|
"eval_gptsynthetic_steps_per_second": 7.252, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 0.86, |
|
"learning_rate": 9.335219243376313e-06, |
|
"loss": 0.4483, |
|
"step": 1810 |
|
}, |
|
{ |
|
"epoch": 0.87, |
|
"learning_rate": 9.325259077137371e-06, |
|
"loss": 0.4317, |
|
"step": 1820 |
|
}, |
|
{ |
|
"epoch": 0.87, |
|
"learning_rate": 9.315230245499886e-06, |
|
"loss": 0.4285, |
|
"step": 1830 |
|
}, |
|
{ |
|
"epoch": 0.88, |
|
"learning_rate": 9.305132907675987e-06, |
|
"loss": 0.4044, |
|
"step": 1840 |
|
}, |
|
{ |
|
"epoch": 0.88, |
|
"learning_rate": 9.294967223965377e-06, |
|
"loss": 0.4418, |
|
"step": 1850 |
|
}, |
|
{ |
|
"epoch": 0.89, |
|
"learning_rate": 9.284733355752775e-06, |
|
"loss": 0.4336, |
|
"step": 1860 |
|
}, |
|
{ |
|
"epoch": 0.89, |
|
"learning_rate": 9.274431465505357e-06, |
|
"loss": 0.4481, |
|
"step": 1870 |
|
}, |
|
{ |
|
"epoch": 0.9, |
|
"learning_rate": 9.264061716770183e-06, |
|
"loss": 0.4294, |
|
"step": 1880 |
|
}, |
|
{ |
|
"epoch": 0.9, |
|
"learning_rate": 9.253624274171595e-06, |
|
"loss": 0.4442, |
|
"step": 1890 |
|
}, |
|
{ |
|
"epoch": 0.91, |
|
"learning_rate": 9.243119303408606e-06, |
|
"loss": 0.4618, |
|
"step": 1900 |
|
}, |
|
{ |
|
"epoch": 0.91, |
|
"learning_rate": 9.232546971252265e-06, |
|
"loss": 0.4286, |
|
"step": 1910 |
|
}, |
|
{ |
|
"epoch": 0.92, |
|
"learning_rate": 9.22190744554302e-06, |
|
"loss": 0.4247, |
|
"step": 1920 |
|
}, |
|
{ |
|
"epoch": 0.92, |
|
"learning_rate": 9.211200895188035e-06, |
|
"loss": 0.4145, |
|
"step": 1930 |
|
}, |
|
{ |
|
"epoch": 0.93, |
|
"learning_rate": 9.200427490158532e-06, |
|
"loss": 0.4019, |
|
"step": 1940 |
|
}, |
|
{ |
|
"epoch": 0.93, |
|
"learning_rate": 9.189587401487073e-06, |
|
"loss": 0.4154, |
|
"step": 1950 |
|
}, |
|
{ |
|
"epoch": 0.94, |
|
"learning_rate": 9.178680801264855e-06, |
|
"loss": 0.4267, |
|
"step": 1960 |
|
}, |
|
{ |
|
"epoch": 0.94, |
|
"learning_rate": 9.167707862638973e-06, |
|
"loss": 0.4344, |
|
"step": 1970 |
|
}, |
|
{ |
|
"epoch": 0.94, |
|
"learning_rate": 9.156668759809678e-06, |
|
"loss": 0.4095, |
|
"step": 1980 |
|
}, |
|
{ |
|
"epoch": 0.95, |
|
"learning_rate": 9.145563668027599e-06, |
|
"loss": 0.4053, |
|
"step": 1990 |
|
}, |
|
{ |
|
"epoch": 0.95, |
|
"learning_rate": 9.134392763590978e-06, |
|
"loss": 0.4244, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.95, |
|
"eval_webgpt_accuracy": 0.585291113381001, |
|
"eval_webgpt_loss": 0.6761194467544556, |
|
"eval_webgpt_runtime": 63.2755, |
|
"eval_webgpt_samples_per_second": 61.888, |
|
"eval_webgpt_steps_per_second": 6.195, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.95, |
|
"eval_hfsummary_accuracy": 0.6690142973732733, |
|
"eval_hfsummary_loss": 0.6501129269599915, |
|
"eval_hfsummary_runtime": 1243.8141, |
|
"eval_hfsummary_samples_per_second": 26.598, |
|
"eval_hfsummary_steps_per_second": 2.66, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.95, |
|
"eval_gptsynthetic_accuracy": 0.9978883861236802, |
|
"eval_gptsynthetic_loss": 0.0076271467842161655, |
|
"eval_gptsynthetic_runtime": 45.8118, |
|
"eval_gptsynthetic_samples_per_second": 72.361, |
|
"eval_gptsynthetic_steps_per_second": 7.247, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.96, |
|
"learning_rate": 9.124282826330867e-06, |
|
"loss": 0.4132, |
|
"step": 2010 |
|
}, |
|
{ |
|
"epoch": 0.96, |
|
"learning_rate": 9.112987367291958e-06, |
|
"loss": 0.4089, |
|
"step": 2020 |
|
}, |
|
{ |
|
"epoch": 0.97, |
|
"learning_rate": 9.101626612761673e-06, |
|
"loss": 0.42, |
|
"step": 2030 |
|
}, |
|
{ |
|
"epoch": 0.97, |
|
"learning_rate": 9.090200743097004e-06, |
|
"loss": 0.3961, |
|
"step": 2040 |
|
}, |
|
{ |
|
"epoch": 0.98, |
|
"learning_rate": 9.078709939688683e-06, |
|
"loss": 0.4139, |
|
"step": 2050 |
|
}, |
|
{ |
|
"epoch": 0.98, |
|
"learning_rate": 9.06715438495829e-06, |
|
"loss": 0.4021, |
|
"step": 2060 |
|
}, |
|
{ |
|
"epoch": 0.99, |
|
"learning_rate": 9.055534262355358e-06, |
|
"loss": 0.4246, |
|
"step": 2070 |
|
}, |
|
{ |
|
"epoch": 0.99, |
|
"learning_rate": 9.043849756354469e-06, |
|
"loss": 0.4233, |
|
"step": 2080 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"learning_rate": 9.032101052452318e-06, |
|
"loss": 0.4121, |
|
"step": 2090 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"learning_rate": 9.020288337164766e-06, |
|
"loss": 0.4299, |
|
"step": 2100 |
|
}, |
|
{ |
|
"epoch": 1.01, |
|
"learning_rate": 9.00841179802389e-06, |
|
"loss": 0.3955, |
|
"step": 2110 |
|
}, |
|
{ |
|
"epoch": 1.01, |
|
"learning_rate": 8.996471623574997e-06, |
|
"loss": 0.4185, |
|
"step": 2120 |
|
}, |
|
{ |
|
"epoch": 1.02, |
|
"learning_rate": 8.984468003373625e-06, |
|
"loss": 0.4086, |
|
"step": 2130 |
|
}, |
|
{ |
|
"epoch": 1.02, |
|
"learning_rate": 8.97240112798255e-06, |
|
"loss": 0.4061, |
|
"step": 2140 |
|
}, |
|
{ |
|
"epoch": 1.03, |
|
"learning_rate": 8.960271188968753e-06, |
|
"loss": 0.3859, |
|
"step": 2150 |
|
}, |
|
{ |
|
"epoch": 1.03, |
|
"learning_rate": 8.948078378900368e-06, |
|
"loss": 0.3951, |
|
"step": 2160 |
|
}, |
|
{ |
|
"epoch": 1.04, |
|
"learning_rate": 8.935822891343647e-06, |
|
"loss": 0.3914, |
|
"step": 2170 |
|
}, |
|
{ |
|
"epoch": 1.04, |
|
"learning_rate": 8.923504920859864e-06, |
|
"loss": 0.3672, |
|
"step": 2180 |
|
}, |
|
{ |
|
"epoch": 1.04, |
|
"learning_rate": 8.911124663002243e-06, |
|
"loss": 0.385, |
|
"step": 2190 |
|
}, |
|
{ |
|
"epoch": 1.05, |
|
"learning_rate": 8.898682314312852e-06, |
|
"loss": 0.3932, |
|
"step": 2200 |
|
}, |
|
{ |
|
"epoch": 1.05, |
|
"eval_webgpt_accuracy": 0.5681818181818182, |
|
"eval_webgpt_loss": 0.6895765662193298, |
|
"eval_webgpt_runtime": 63.5181, |
|
"eval_webgpt_samples_per_second": 61.652, |
|
"eval_webgpt_steps_per_second": 6.171, |
|
"step": 2200 |
|
}, |
|
{ |
|
"epoch": 1.05, |
|
"eval_hfsummary_accuracy": 0.6638757065562373, |
|
"eval_hfsummary_loss": 0.6736522912979126, |
|
"eval_hfsummary_runtime": 1274.8625, |
|
"eval_hfsummary_samples_per_second": 25.95, |
|
"eval_hfsummary_steps_per_second": 2.596, |
|
"step": 2200 |
|
}, |
|
{ |
|
"epoch": 1.05, |
|
"eval_gptsynthetic_accuracy": 0.9984917043740573, |
|
"eval_gptsynthetic_loss": 0.006684631574898958, |
|
"eval_gptsynthetic_runtime": 46.0527, |
|
"eval_gptsynthetic_samples_per_second": 71.983, |
|
"eval_gptsynthetic_steps_per_second": 7.209, |
|
"step": 2200 |
|
}, |
|
{ |
|
"epoch": 1.05, |
|
"learning_rate": 8.886178072319464e-06, |
|
"loss": 0.3868, |
|
"step": 2210 |
|
}, |
|
{ |
|
"epoch": 1.06, |
|
"learning_rate": 8.873612135532451e-06, |
|
"loss": 0.3712, |
|
"step": 2220 |
|
}, |
|
{ |
|
"epoch": 1.06, |
|
"learning_rate": 8.860984703441608e-06, |
|
"loss": 0.4153, |
|
"step": 2230 |
|
}, |
|
{ |
|
"epoch": 1.07, |
|
"learning_rate": 8.848295976512996e-06, |
|
"loss": 0.4364, |
|
"step": 2240 |
|
}, |
|
{ |
|
"epoch": 1.07, |
|
"learning_rate": 8.835546156185765e-06, |
|
"loss": 0.3733, |
|
"step": 2250 |
|
}, |
|
{ |
|
"epoch": 1.08, |
|
"learning_rate": 8.822735444868941e-06, |
|
"loss": 0.389, |
|
"step": 2260 |
|
}, |
|
{ |
|
"epoch": 1.08, |
|
"learning_rate": 8.809864045938227e-06, |
|
"loss": 0.3869, |
|
"step": 2270 |
|
}, |
|
{ |
|
"epoch": 1.09, |
|
"learning_rate": 8.796932163732768e-06, |
|
"loss": 0.3782, |
|
"step": 2280 |
|
}, |
|
{ |
|
"epoch": 1.09, |
|
"learning_rate": 8.783940003551902e-06, |
|
"loss": 0.4087, |
|
"step": 2290 |
|
}, |
|
{ |
|
"epoch": 1.1, |
|
"learning_rate": 8.770887771651916e-06, |
|
"loss": 0.3849, |
|
"step": 2300 |
|
}, |
|
{ |
|
"epoch": 1.1, |
|
"learning_rate": 8.757775675242757e-06, |
|
"loss": 0.3848, |
|
"step": 2310 |
|
}, |
|
{ |
|
"epoch": 1.11, |
|
"learning_rate": 8.744603922484746e-06, |
|
"loss": 0.4036, |
|
"step": 2320 |
|
}, |
|
{ |
|
"epoch": 1.11, |
|
"learning_rate": 8.731372722485276e-06, |
|
"loss": 0.3931, |
|
"step": 2330 |
|
}, |
|
{ |
|
"epoch": 1.12, |
|
"learning_rate": 8.718082285295496e-06, |
|
"loss": 0.3909, |
|
"step": 2340 |
|
}, |
|
{ |
|
"epoch": 1.12, |
|
"learning_rate": 8.70473282190696e-06, |
|
"loss": 0.3912, |
|
"step": 2350 |
|
}, |
|
{ |
|
"epoch": 1.13, |
|
"learning_rate": 8.691324544248303e-06, |
|
"loss": 0.3874, |
|
"step": 2360 |
|
}, |
|
{ |
|
"epoch": 1.13, |
|
"learning_rate": 8.677857665181854e-06, |
|
"loss": 0.3878, |
|
"step": 2370 |
|
}, |
|
{ |
|
"epoch": 1.14, |
|
"learning_rate": 8.664332398500269e-06, |
|
"loss": 0.4063, |
|
"step": 2380 |
|
}, |
|
{ |
|
"epoch": 1.14, |
|
"learning_rate": 8.650748958923128e-06, |
|
"loss": 0.3794, |
|
"step": 2390 |
|
}, |
|
{ |
|
"epoch": 1.15, |
|
"learning_rate": 8.637107562093537e-06, |
|
"loss": 0.3514, |
|
"step": 2400 |
|
}, |
|
{ |
|
"epoch": 1.15, |
|
"eval_webgpt_accuracy": 0.5840143003064351, |
|
"eval_webgpt_loss": 0.6855098605155945, |
|
"eval_webgpt_runtime": 63.4383, |
|
"eval_webgpt_samples_per_second": 61.729, |
|
"eval_webgpt_steps_per_second": 6.179, |
|
"step": 2400 |
|
}, |
|
{ |
|
"epoch": 1.15, |
|
"eval_hfsummary_accuracy": 0.6732158510413203, |
|
"eval_hfsummary_loss": 0.6792841553688049, |
|
"eval_hfsummary_runtime": 1256.3167, |
|
"eval_hfsummary_samples_per_second": 26.333, |
|
"eval_hfsummary_steps_per_second": 2.634, |
|
"step": 2400 |
|
}, |
|
{ |
|
"epoch": 1.15, |
|
"eval_gptsynthetic_accuracy": 0.9978883861236802, |
|
"eval_gptsynthetic_loss": 0.00653717340901494, |
|
"eval_gptsynthetic_runtime": 45.9265, |
|
"eval_gptsynthetic_samples_per_second": 72.181, |
|
"eval_gptsynthetic_steps_per_second": 7.229, |
|
"step": 2400 |
|
}, |
|
{ |
|
"epoch": 1.15, |
|
"learning_rate": 8.623408424574695e-06, |
|
"loss": 0.3742, |
|
"step": 2410 |
|
}, |
|
{ |
|
"epoch": 1.15, |
|
"learning_rate": 8.609651763846462e-06, |
|
"loss": 0.3696, |
|
"step": 2420 |
|
}, |
|
{ |
|
"epoch": 1.16, |
|
"learning_rate": 8.595837798301908e-06, |
|
"loss": 0.3943, |
|
"step": 2430 |
|
}, |
|
{ |
|
"epoch": 1.16, |
|
"learning_rate": 8.581966747243835e-06, |
|
"loss": 0.3928, |
|
"step": 2440 |
|
}, |
|
{ |
|
"epoch": 1.17, |
|
"learning_rate": 8.568038830881305e-06, |
|
"loss": 0.3753, |
|
"step": 2450 |
|
}, |
|
{ |
|
"epoch": 1.17, |
|
"learning_rate": 8.554054270326144e-06, |
|
"loss": 0.3594, |
|
"step": 2460 |
|
}, |
|
{ |
|
"epoch": 1.18, |
|
"learning_rate": 8.54001328758943e-06, |
|
"loss": 0.3674, |
|
"step": 2470 |
|
}, |
|
{ |
|
"epoch": 1.18, |
|
"learning_rate": 8.525916105577967e-06, |
|
"loss": 0.3714, |
|
"step": 2480 |
|
}, |
|
{ |
|
"epoch": 1.19, |
|
"learning_rate": 8.511762948090745e-06, |
|
"loss": 0.3937, |
|
"step": 2490 |
|
}, |
|
{ |
|
"epoch": 1.19, |
|
"learning_rate": 8.49755403981539e-06, |
|
"loss": 0.3848, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 1.2, |
|
"learning_rate": 8.4832896063246e-06, |
|
"loss": 0.3709, |
|
"step": 2510 |
|
}, |
|
{ |
|
"epoch": 1.2, |
|
"learning_rate": 8.46896987407256e-06, |
|
"loss": 0.4056, |
|
"step": 2520 |
|
}, |
|
{ |
|
"epoch": 1.21, |
|
"learning_rate": 8.45459507039134e-06, |
|
"loss": 0.3794, |
|
"step": 2530 |
|
}, |
|
{ |
|
"epoch": 1.21, |
|
"learning_rate": 8.440165423487302e-06, |
|
"loss": 0.3597, |
|
"step": 2540 |
|
}, |
|
{ |
|
"epoch": 1.22, |
|
"learning_rate": 8.425681162437464e-06, |
|
"loss": 0.3773, |
|
"step": 2550 |
|
}, |
|
{ |
|
"epoch": 1.22, |
|
"learning_rate": 8.41114251718587e-06, |
|
"loss": 0.3678, |
|
"step": 2560 |
|
}, |
|
{ |
|
"epoch": 1.23, |
|
"learning_rate": 8.396549718539934e-06, |
|
"loss": 0.3831, |
|
"step": 2570 |
|
}, |
|
{ |
|
"epoch": 1.23, |
|
"learning_rate": 8.381902998166782e-06, |
|
"loss": 0.3873, |
|
"step": 2580 |
|
}, |
|
{ |
|
"epoch": 1.24, |
|
"learning_rate": 8.367202588589569e-06, |
|
"loss": 0.3775, |
|
"step": 2590 |
|
}, |
|
{ |
|
"epoch": 1.24, |
|
"learning_rate": 8.352448723183792e-06, |
|
"loss": 0.3936, |
|
"step": 2600 |
|
}, |
|
{ |
|
"epoch": 1.24, |
|
"eval_webgpt_accuracy": 0.5760980592441267, |
|
"eval_webgpt_loss": 0.6807132959365845, |
|
"eval_webgpt_runtime": 68.3724, |
|
"eval_webgpt_samples_per_second": 57.275, |
|
"eval_webgpt_steps_per_second": 5.733, |
|
"step": 2600 |
|
}, |
|
{ |
|
"epoch": 1.24, |
|
"eval_hfsummary_accuracy": 0.6699815615270683, |
|
"eval_hfsummary_loss": 0.6664829254150391, |
|
"eval_hfsummary_runtime": 1323.7739, |
|
"eval_hfsummary_samples_per_second": 24.991, |
|
"eval_hfsummary_steps_per_second": 2.5, |
|
"step": 2600 |
|
}, |
|
{ |
|
"epoch": 1.24, |
|
"eval_gptsynthetic_accuracy": 0.9984917043740573, |
|
"eval_gptsynthetic_loss": 0.006850704550743103, |
|
"eval_gptsynthetic_runtime": 48.9916, |
|
"eval_gptsynthetic_samples_per_second": 67.665, |
|
"eval_gptsynthetic_steps_per_second": 6.777, |
|
"step": 2600 |
|
}, |
|
{ |
|
"epoch": 1.25, |
|
"learning_rate": 8.337641636173582e-06, |
|
"loss": 0.3647, |
|
"step": 2610 |
|
}, |
|
{ |
|
"epoch": 1.25, |
|
"learning_rate": 8.322781562627987e-06, |
|
"loss": 0.3815, |
|
"step": 2620 |
|
}, |
|
{ |
|
"epoch": 1.25, |
|
"learning_rate": 8.307868738457236e-06, |
|
"loss": 0.3979, |
|
"step": 2630 |
|
}, |
|
{ |
|
"epoch": 1.26, |
|
"learning_rate": 8.292903400409005e-06, |
|
"loss": 0.3685, |
|
"step": 2640 |
|
}, |
|
{ |
|
"epoch": 1.26, |
|
"learning_rate": 8.27788578606464e-06, |
|
"loss": 0.3498, |
|
"step": 2650 |
|
}, |
|
{ |
|
"epoch": 1.27, |
|
"learning_rate": 8.262816133835412e-06, |
|
"loss": 0.3641, |
|
"step": 2660 |
|
}, |
|
{ |
|
"epoch": 1.27, |
|
"learning_rate": 8.247694682958699e-06, |
|
"loss": 0.4006, |
|
"step": 2670 |
|
}, |
|
{ |
|
"epoch": 1.28, |
|
"learning_rate": 8.232521673494216e-06, |
|
"loss": 0.3857, |
|
"step": 2680 |
|
}, |
|
{ |
|
"epoch": 1.28, |
|
"learning_rate": 8.217297346320192e-06, |
|
"loss": 0.3838, |
|
"step": 2690 |
|
}, |
|
{ |
|
"epoch": 1.29, |
|
"learning_rate": 8.202021943129548e-06, |
|
"loss": 0.3979, |
|
"step": 2700 |
|
}, |
|
{ |
|
"epoch": 1.29, |
|
"learning_rate": 8.186695706426053e-06, |
|
"loss": 0.3589, |
|
"step": 2710 |
|
}, |
|
{ |
|
"epoch": 1.3, |
|
"learning_rate": 8.17131887952049e-06, |
|
"loss": 0.3853, |
|
"step": 2720 |
|
}, |
|
{ |
|
"epoch": 1.3, |
|
"learning_rate": 8.155891706526773e-06, |
|
"loss": 0.3698, |
|
"step": 2730 |
|
}, |
|
{ |
|
"epoch": 1.31, |
|
"learning_rate": 8.140414432358094e-06, |
|
"loss": 0.3637, |
|
"step": 2740 |
|
}, |
|
{ |
|
"epoch": 1.31, |
|
"learning_rate": 8.124887302723014e-06, |
|
"loss": 0.3742, |
|
"step": 2750 |
|
}, |
|
{ |
|
"epoch": 1.32, |
|
"learning_rate": 8.109310564121577e-06, |
|
"loss": 0.3849, |
|
"step": 2760 |
|
}, |
|
{ |
|
"epoch": 1.32, |
|
"learning_rate": 8.09368446384139e-06, |
|
"loss": 0.3721, |
|
"step": 2770 |
|
}, |
|
{ |
|
"epoch": 1.33, |
|
"learning_rate": 8.078009249953693e-06, |
|
"loss": 0.3818, |
|
"step": 2780 |
|
}, |
|
{ |
|
"epoch": 1.33, |
|
"learning_rate": 8.062285171309437e-06, |
|
"loss": 0.384, |
|
"step": 2790 |
|
}, |
|
{ |
|
"epoch": 1.34, |
|
"learning_rate": 8.046512477535315e-06, |
|
"loss": 0.3567, |
|
"step": 2800 |
|
}, |
|
{ |
|
"epoch": 1.34, |
|
"eval_webgpt_accuracy": 0.5822267620020429, |
|
"eval_webgpt_loss": 0.6815086007118225, |
|
"eval_webgpt_runtime": 68.1221, |
|
"eval_webgpt_samples_per_second": 57.485, |
|
"eval_webgpt_steps_per_second": 5.754, |
|
"step": 2800 |
|
}, |
|
{ |
|
"epoch": 1.34, |
|
"eval_hfsummary_accuracy": 0.6710697337000877, |
|
"eval_hfsummary_loss": 0.6821005940437317, |
|
"eval_hfsummary_runtime": 1319.5773, |
|
"eval_hfsummary_samples_per_second": 25.071, |
|
"eval_hfsummary_steps_per_second": 2.508, |
|
"step": 2800 |
|
}, |
|
{ |
|
"epoch": 1.34, |
|
"eval_gptsynthetic_accuracy": 0.9978883861236802, |
|
"eval_gptsynthetic_loss": 0.005701866466552019, |
|
"eval_gptsynthetic_runtime": 48.8957, |
|
"eval_gptsynthetic_samples_per_second": 67.797, |
|
"eval_gptsynthetic_steps_per_second": 6.79, |
|
"step": 2800 |
|
}, |
|
{ |
|
"epoch": 1.34, |
|
"learning_rate": 8.030691419029806e-06, |
|
"loss": 0.3813, |
|
"step": 2810 |
|
}, |
|
{ |
|
"epoch": 1.35, |
|
"learning_rate": 8.014822246959203e-06, |
|
"loss": 0.3645, |
|
"step": 2820 |
|
}, |
|
{ |
|
"epoch": 1.35, |
|
"learning_rate": 7.998905213253627e-06, |
|
"loss": 0.3752, |
|
"step": 2830 |
|
}, |
|
{ |
|
"epoch": 1.35, |
|
"learning_rate": 7.982940570603014e-06, |
|
"loss": 0.4075, |
|
"step": 2840 |
|
}, |
|
{ |
|
"epoch": 1.36, |
|
"learning_rate": 7.966928572453123e-06, |
|
"loss": 0.3823, |
|
"step": 2850 |
|
}, |
|
{ |
|
"epoch": 1.36, |
|
"learning_rate": 7.950869473001494e-06, |
|
"loss": 0.3776, |
|
"step": 2860 |
|
}, |
|
{ |
|
"epoch": 1.37, |
|
"learning_rate": 7.93476352719343e-06, |
|
"loss": 0.3689, |
|
"step": 2870 |
|
}, |
|
{ |
|
"epoch": 1.37, |
|
"learning_rate": 7.91861099071793e-06, |
|
"loss": 0.3883, |
|
"step": 2880 |
|
}, |
|
{ |
|
"epoch": 1.38, |
|
"learning_rate": 7.902412120003649e-06, |
|
"loss": 0.3736, |
|
"step": 2890 |
|
}, |
|
{ |
|
"epoch": 1.38, |
|
"learning_rate": 7.886167172214814e-06, |
|
"loss": 0.3363, |
|
"step": 2900 |
|
}, |
|
{ |
|
"epoch": 1.39, |
|
"learning_rate": 7.86987640524715e-06, |
|
"loss": 0.3866, |
|
"step": 2910 |
|
}, |
|
{ |
|
"epoch": 1.39, |
|
"learning_rate": 7.853540077723779e-06, |
|
"loss": 0.3717, |
|
"step": 2920 |
|
}, |
|
{ |
|
"epoch": 1.4, |
|
"learning_rate": 7.837158448991112e-06, |
|
"loss": 0.3665, |
|
"step": 2930 |
|
}, |
|
{ |
|
"epoch": 1.4, |
|
"learning_rate": 7.820731779114746e-06, |
|
"loss": 0.3814, |
|
"step": 2940 |
|
}, |
|
{ |
|
"epoch": 1.41, |
|
"learning_rate": 7.804260328875322e-06, |
|
"loss": 0.3703, |
|
"step": 2950 |
|
}, |
|
{ |
|
"epoch": 1.41, |
|
"learning_rate": 7.787744359764388e-06, |
|
"loss": 0.3896, |
|
"step": 2960 |
|
}, |
|
{ |
|
"epoch": 1.42, |
|
"learning_rate": 7.77118413398025e-06, |
|
"loss": 0.3566, |
|
"step": 2970 |
|
}, |
|
{ |
|
"epoch": 1.42, |
|
"learning_rate": 7.754579914423806e-06, |
|
"loss": 0.3743, |
|
"step": 2980 |
|
}, |
|
{ |
|
"epoch": 1.43, |
|
"learning_rate": 7.737931964694378e-06, |
|
"loss": 0.3657, |
|
"step": 2990 |
|
}, |
|
{ |
|
"epoch": 1.43, |
|
"learning_rate": 7.721240549085523e-06, |
|
"loss": 0.3956, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 1.43, |
|
"eval_webgpt_accuracy": 0.5799284984678243, |
|
"eval_webgpt_loss": 0.6848716139793396, |
|
"eval_webgpt_runtime": 67.9306, |
|
"eval_webgpt_samples_per_second": 57.647, |
|
"eval_webgpt_steps_per_second": 5.771, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 1.43, |
|
"eval_hfsummary_accuracy": 0.6674122661185503, |
|
"eval_hfsummary_loss": 0.6731451153755188, |
|
"eval_hfsummary_runtime": 1273.6757, |
|
"eval_hfsummary_samples_per_second": 25.974, |
|
"eval_hfsummary_steps_per_second": 2.598, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 1.43, |
|
"eval_gptsynthetic_accuracy": 0.9984917043740573, |
|
"eval_gptsynthetic_loss": 0.005726581439375877, |
|
"eval_gptsynthetic_runtime": 48.6337, |
|
"eval_gptsynthetic_samples_per_second": 68.163, |
|
"eval_gptsynthetic_steps_per_second": 6.827, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 1.44, |
|
"learning_rate": 7.704505932580835e-06, |
|
"loss": 0.3617, |
|
"step": 3010 |
|
}, |
|
{ |
|
"epoch": 1.44, |
|
"learning_rate": 7.687728380849746e-06, |
|
"loss": 0.3944, |
|
"step": 3020 |
|
}, |
|
{ |
|
"epoch": 1.45, |
|
"learning_rate": 7.6709081602433e-06, |
|
"loss": 0.347, |
|
"step": 3030 |
|
}, |
|
{ |
|
"epoch": 1.45, |
|
"learning_rate": 7.654045537789929e-06, |
|
"loss": 0.3657, |
|
"step": 3040 |
|
}, |
|
{ |
|
"epoch": 1.46, |
|
"learning_rate": 7.637140781191214e-06, |
|
"loss": 0.3874, |
|
"step": 3050 |
|
}, |
|
{ |
|
"epoch": 1.46, |
|
"learning_rate": 7.6201941588176355e-06, |
|
"loss": 0.3875, |
|
"step": 3060 |
|
}, |
|
{ |
|
"epoch": 1.46, |
|
"learning_rate": 7.603205939704308e-06, |
|
"loss": 0.371, |
|
"step": 3070 |
|
}, |
|
{ |
|
"epoch": 1.47, |
|
"learning_rate": 7.586176393546714e-06, |
|
"loss": 0.4006, |
|
"step": 3080 |
|
}, |
|
{ |
|
"epoch": 1.47, |
|
"learning_rate": 7.569105790696423e-06, |
|
"loss": 0.3721, |
|
"step": 3090 |
|
}, |
|
{ |
|
"epoch": 1.48, |
|
"learning_rate": 7.551994402156794e-06, |
|
"loss": 0.3825, |
|
"step": 3100 |
|
}, |
|
{ |
|
"epoch": 1.48, |
|
"learning_rate": 7.5348424995786814e-06, |
|
"loss": 0.3688, |
|
"step": 3110 |
|
}, |
|
{ |
|
"epoch": 1.49, |
|
"learning_rate": 7.517650355256116e-06, |
|
"loss": 0.3698, |
|
"step": 3120 |
|
}, |
|
{ |
|
"epoch": 1.49, |
|
"learning_rate": 7.500418242121982e-06, |
|
"loss": 0.3623, |
|
"step": 3130 |
|
}, |
|
{ |
|
"epoch": 1.5, |
|
"learning_rate": 7.483146433743686e-06, |
|
"loss": 0.3739, |
|
"step": 3140 |
|
}, |
|
{ |
|
"epoch": 1.5, |
|
"learning_rate": 7.465835204318822e-06, |
|
"loss": 0.3708, |
|
"step": 3150 |
|
}, |
|
{ |
|
"epoch": 1.51, |
|
"learning_rate": 7.448484828670799e-06, |
|
"loss": 0.3344, |
|
"step": 3160 |
|
}, |
|
{ |
|
"epoch": 1.51, |
|
"learning_rate": 7.431095582244495e-06, |
|
"loss": 0.3548, |
|
"step": 3170 |
|
}, |
|
{ |
|
"epoch": 1.52, |
|
"learning_rate": 7.413667741101881e-06, |
|
"loss": 0.3702, |
|
"step": 3180 |
|
}, |
|
{ |
|
"epoch": 1.52, |
|
"learning_rate": 7.3962015819176324e-06, |
|
"loss": 0.3361, |
|
"step": 3190 |
|
}, |
|
{ |
|
"epoch": 1.53, |
|
"learning_rate": 7.378697381974744e-06, |
|
"loss": 0.4019, |
|
"step": 3200 |
|
}, |
|
{ |
|
"epoch": 1.53, |
|
"eval_webgpt_accuracy": 0.5891215526046987, |
|
"eval_webgpt_loss": 0.6808570623397827, |
|
"eval_webgpt_runtime": 63.3758, |
|
"eval_webgpt_samples_per_second": 61.79, |
|
"eval_webgpt_steps_per_second": 6.185, |
|
"step": 3200 |
|
}, |
|
{ |
|
"epoch": 1.53, |
|
"eval_hfsummary_accuracy": 0.6569839494604479, |
|
"eval_hfsummary_loss": 0.692561686038971, |
|
"eval_hfsummary_runtime": 1242.943, |
|
"eval_hfsummary_samples_per_second": 26.617, |
|
"eval_hfsummary_steps_per_second": 2.662, |
|
"step": 3200 |
|
}, |
|
{ |
|
"epoch": 1.53, |
|
"eval_gptsynthetic_accuracy": 0.9987933634992459, |
|
"eval_gptsynthetic_loss": 0.005168143659830093, |
|
"eval_gptsynthetic_runtime": 45.8689, |
|
"eval_gptsynthetic_samples_per_second": 72.271, |
|
"eval_gptsynthetic_steps_per_second": 7.238, |
|
"step": 3200 |
|
}, |
|
{ |
|
"epoch": 1.53, |
|
"learning_rate": 7.361155419160121e-06, |
|
"loss": 0.3573, |
|
"step": 3210 |
|
}, |
|
{ |
|
"epoch": 1.54, |
|
"learning_rate": 7.343575971960171e-06, |
|
"loss": 0.3677, |
|
"step": 3220 |
|
}, |
|
{ |
|
"epoch": 1.54, |
|
"learning_rate": 7.325959319456386e-06, |
|
"loss": 0.365, |
|
"step": 3230 |
|
}, |
|
{ |
|
"epoch": 1.55, |
|
"learning_rate": 7.308305741320907e-06, |
|
"loss": 0.3241, |
|
"step": 3240 |
|
}, |
|
{ |
|
"epoch": 1.55, |
|
"learning_rate": 7.290615517812084e-06, |
|
"loss": 0.3639, |
|
"step": 3250 |
|
}, |
|
{ |
|
"epoch": 1.56, |
|
"learning_rate": 7.27288892977003e-06, |
|
"loss": 0.3535, |
|
"step": 3260 |
|
}, |
|
{ |
|
"epoch": 1.56, |
|
"learning_rate": 7.255126258612163e-06, |
|
"loss": 0.364, |
|
"step": 3270 |
|
}, |
|
{ |
|
"epoch": 1.56, |
|
"learning_rate": 7.237327786328732e-06, |
|
"loss": 0.3728, |
|
"step": 3280 |
|
}, |
|
{ |
|
"epoch": 1.57, |
|
"learning_rate": 7.219493795478347e-06, |
|
"loss": 0.3598, |
|
"step": 3290 |
|
}, |
|
{ |
|
"epoch": 1.57, |
|
"learning_rate": 7.201624569183492e-06, |
|
"loss": 0.3749, |
|
"step": 3300 |
|
}, |
|
{ |
|
"epoch": 1.58, |
|
"learning_rate": 7.183720391126028e-06, |
|
"loss": 0.3361, |
|
"step": 3310 |
|
}, |
|
{ |
|
"epoch": 1.58, |
|
"learning_rate": 7.16578154554269e-06, |
|
"loss": 0.3383, |
|
"step": 3320 |
|
}, |
|
{ |
|
"epoch": 1.59, |
|
"learning_rate": 7.147808317220578e-06, |
|
"loss": 0.3424, |
|
"step": 3330 |
|
}, |
|
{ |
|
"epoch": 1.59, |
|
"learning_rate": 7.12980099149263e-06, |
|
"loss": 0.3627, |
|
"step": 3340 |
|
}, |
|
{ |
|
"epoch": 1.6, |
|
"learning_rate": 7.111759854233096e-06, |
|
"loss": 0.3498, |
|
"step": 3350 |
|
}, |
|
{ |
|
"epoch": 1.6, |
|
"learning_rate": 7.093685191853003e-06, |
|
"loss": 0.3838, |
|
"step": 3360 |
|
}, |
|
{ |
|
"epoch": 1.61, |
|
"learning_rate": 7.075577291295599e-06, |
|
"loss": 0.3638, |
|
"step": 3370 |
|
}, |
|
{ |
|
"epoch": 1.61, |
|
"learning_rate": 7.0574364400318014e-06, |
|
"loss": 0.3737, |
|
"step": 3380 |
|
}, |
|
{ |
|
"epoch": 1.62, |
|
"learning_rate": 7.0392629260556425e-06, |
|
"loss": 0.367, |
|
"step": 3390 |
|
}, |
|
{ |
|
"epoch": 1.62, |
|
"learning_rate": 7.021057037879681e-06, |
|
"loss": 0.3778, |
|
"step": 3400 |
|
}, |
|
{ |
|
"epoch": 1.62, |
|
"eval_webgpt_accuracy": 0.5845250255362615, |
|
"eval_webgpt_loss": 0.6846094727516174, |
|
"eval_webgpt_runtime": 63.3427, |
|
"eval_webgpt_samples_per_second": 61.822, |
|
"eval_webgpt_steps_per_second": 6.189, |
|
"step": 3400 |
|
}, |
|
{ |
|
"epoch": 1.62, |
|
"eval_hfsummary_accuracy": 0.6690445243780794, |
|
"eval_hfsummary_loss": 0.6792688369750977, |
|
"eval_hfsummary_runtime": 1245.6911, |
|
"eval_hfsummary_samples_per_second": 26.558, |
|
"eval_hfsummary_steps_per_second": 2.656, |
|
"step": 3400 |
|
}, |
|
{ |
|
"epoch": 1.62, |
|
"eval_gptsynthetic_accuracy": 0.9984917043740573, |
|
"eval_gptsynthetic_loss": 0.005154031794518232, |
|
"eval_gptsynthetic_runtime": 45.8488, |
|
"eval_gptsynthetic_samples_per_second": 72.303, |
|
"eval_gptsynthetic_steps_per_second": 7.241, |
|
"step": 3400 |
|
}, |
|
{ |
|
"epoch": 1.63, |
|
"learning_rate": 7.002819064530437e-06, |
|
"loss": 0.3706, |
|
"step": 3410 |
|
}, |
|
{ |
|
"epoch": 1.63, |
|
"learning_rate": 6.984549295543795e-06, |
|
"loss": 0.3743, |
|
"step": 3420 |
|
}, |
|
{ |
|
"epoch": 1.64, |
|
"learning_rate": 6.966248020960404e-06, |
|
"loss": 0.3613, |
|
"step": 3430 |
|
}, |
|
{ |
|
"epoch": 1.64, |
|
"learning_rate": 6.947915531321085e-06, |
|
"loss": 0.3375, |
|
"step": 3440 |
|
}, |
|
{ |
|
"epoch": 1.65, |
|
"learning_rate": 6.929552117662211e-06, |
|
"loss": 0.3515, |
|
"step": 3450 |
|
}, |
|
{ |
|
"epoch": 1.65, |
|
"learning_rate": 6.911158071511084e-06, |
|
"loss": 0.357, |
|
"step": 3460 |
|
}, |
|
{ |
|
"epoch": 1.66, |
|
"learning_rate": 6.892733684881313e-06, |
|
"loss": 0.3267, |
|
"step": 3470 |
|
}, |
|
{ |
|
"epoch": 1.66, |
|
"learning_rate": 6.874279250268174e-06, |
|
"loss": 0.3562, |
|
"step": 3480 |
|
}, |
|
{ |
|
"epoch": 1.67, |
|
"learning_rate": 6.85579506064397e-06, |
|
"loss": 0.361, |
|
"step": 3490 |
|
}, |
|
{ |
|
"epoch": 1.67, |
|
"learning_rate": 6.837281409453376e-06, |
|
"loss": 0.3452, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 1.67, |
|
"learning_rate": 6.818738590608782e-06, |
|
"loss": 0.348, |
|
"step": 3510 |
|
}, |
|
{ |
|
"epoch": 1.68, |
|
"learning_rate": 6.800166898485632e-06, |
|
"loss": 0.3496, |
|
"step": 3520 |
|
}, |
|
{ |
|
"epoch": 1.68, |
|
"learning_rate": 6.781566627917738e-06, |
|
"loss": 0.3548, |
|
"step": 3530 |
|
}, |
|
{ |
|
"epoch": 1.69, |
|
"learning_rate": 6.762938074192615e-06, |
|
"loss": 0.3302, |
|
"step": 3540 |
|
}, |
|
{ |
|
"epoch": 1.69, |
|
"learning_rate": 6.744281533046782e-06, |
|
"loss": 0.3447, |
|
"step": 3550 |
|
}, |
|
{ |
|
"epoch": 1.7, |
|
"learning_rate": 6.725597300661071e-06, |
|
"loss": 0.3615, |
|
"step": 3560 |
|
}, |
|
{ |
|
"epoch": 1.7, |
|
"learning_rate": 6.706885673655924e-06, |
|
"loss": 0.3263, |
|
"step": 3570 |
|
}, |
|
{ |
|
"epoch": 1.71, |
|
"learning_rate": 6.688146949086688e-06, |
|
"loss": 0.3258, |
|
"step": 3580 |
|
}, |
|
{ |
|
"epoch": 1.71, |
|
"learning_rate": 6.669381424438892e-06, |
|
"loss": 0.3483, |
|
"step": 3590 |
|
}, |
|
{ |
|
"epoch": 1.72, |
|
"learning_rate": 6.650589397623528e-06, |
|
"loss": 0.3643, |
|
"step": 3600 |
|
}, |
|
{ |
|
"epoch": 1.72, |
|
"eval_webgpt_accuracy": 0.5893769152196119, |
|
"eval_webgpt_loss": 0.6877950429916382, |
|
"eval_webgpt_runtime": 63.3493, |
|
"eval_webgpt_samples_per_second": 61.816, |
|
"eval_webgpt_steps_per_second": 6.188, |
|
"step": 3600 |
|
}, |
|
{ |
|
"epoch": 1.72, |
|
"eval_hfsummary_accuracy": 0.6664147749599492, |
|
"eval_hfsummary_loss": 0.7294792532920837, |
|
"eval_hfsummary_runtime": 1242.1345, |
|
"eval_hfsummary_samples_per_second": 26.634, |
|
"eval_hfsummary_steps_per_second": 2.664, |
|
"step": 3600 |
|
}, |
|
{ |
|
"epoch": 1.72, |
|
"eval_gptsynthetic_accuracy": 0.9993966817496229, |
|
"eval_gptsynthetic_loss": 0.005264561623334885, |
|
"eval_gptsynthetic_runtime": 45.9664, |
|
"eval_gptsynthetic_samples_per_second": 72.118, |
|
"eval_gptsynthetic_steps_per_second": 7.223, |
|
"step": 3600 |
|
}, |
|
{ |
|
"epoch": 1.72, |
|
"learning_rate": 6.6317711669723265e-06, |
|
"loss": 0.3443, |
|
"step": 3610 |
|
}, |
|
{ |
|
"epoch": 1.73, |
|
"learning_rate": 6.61292703123301e-06, |
|
"loss": 0.3539, |
|
"step": 3620 |
|
}, |
|
{ |
|
"epoch": 1.73, |
|
"learning_rate": 6.594057289564557e-06, |
|
"loss": 0.3123, |
|
"step": 3630 |
|
}, |
|
{ |
|
"epoch": 1.74, |
|
"learning_rate": 6.575162241532453e-06, |
|
"loss": 0.3047, |
|
"step": 3640 |
|
}, |
|
{ |
|
"epoch": 1.74, |
|
"learning_rate": 6.556242187103934e-06, |
|
"loss": 0.3687, |
|
"step": 3650 |
|
}, |
|
{ |
|
"epoch": 1.75, |
|
"learning_rate": 6.53729742664322e-06, |
|
"loss": 0.3652, |
|
"step": 3660 |
|
}, |
|
{ |
|
"epoch": 1.75, |
|
"learning_rate": 6.518328260906753e-06, |
|
"loss": 0.3493, |
|
"step": 3670 |
|
}, |
|
{ |
|
"epoch": 1.76, |
|
"learning_rate": 6.49933499103842e-06, |
|
"loss": 0.3637, |
|
"step": 3680 |
|
}, |
|
{ |
|
"epoch": 1.76, |
|
"learning_rate": 6.480317918564767e-06, |
|
"loss": 0.3464, |
|
"step": 3690 |
|
}, |
|
{ |
|
"epoch": 1.77, |
|
"learning_rate": 6.4612773453902244e-06, |
|
"loss": 0.3376, |
|
"step": 3700 |
|
}, |
|
{ |
|
"epoch": 1.77, |
|
"learning_rate": 6.4422135737923e-06, |
|
"loss": 0.3473, |
|
"step": 3710 |
|
}, |
|
{ |
|
"epoch": 1.77, |
|
"learning_rate": 6.423126906416789e-06, |
|
"loss": 0.3356, |
|
"step": 3720 |
|
}, |
|
{ |
|
"epoch": 1.78, |
|
"learning_rate": 6.40401764627297e-06, |
|
"loss": 0.3374, |
|
"step": 3730 |
|
}, |
|
{ |
|
"epoch": 1.78, |
|
"learning_rate": 6.384886096728789e-06, |
|
"loss": 0.3544, |
|
"step": 3740 |
|
}, |
|
{ |
|
"epoch": 1.79, |
|
"learning_rate": 6.365732561506044e-06, |
|
"loss": 0.3361, |
|
"step": 3750 |
|
}, |
|
{ |
|
"epoch": 1.79, |
|
"learning_rate": 6.346557344675568e-06, |
|
"loss": 0.3689, |
|
"step": 3760 |
|
}, |
|
{ |
|
"epoch": 1.8, |
|
"learning_rate": 6.327360750652402e-06, |
|
"loss": 0.359, |
|
"step": 3770 |
|
}, |
|
{ |
|
"epoch": 1.8, |
|
"learning_rate": 6.308143084190954e-06, |
|
"loss": 0.3423, |
|
"step": 3780 |
|
}, |
|
{ |
|
"epoch": 1.81, |
|
"learning_rate": 6.2889046503801685e-06, |
|
"loss": 0.3392, |
|
"step": 3790 |
|
}, |
|
{ |
|
"epoch": 1.81, |
|
"learning_rate": 6.269645754638682e-06, |
|
"loss": 0.342, |
|
"step": 3800 |
|
}, |
|
{ |
|
"epoch": 1.81, |
|
"eval_webgpt_accuracy": 0.5842696629213483, |
|
"eval_webgpt_loss": 0.6836753487586975, |
|
"eval_webgpt_runtime": 63.4162, |
|
"eval_webgpt_samples_per_second": 61.751, |
|
"eval_webgpt_steps_per_second": 6.181, |
|
"step": 3800 |
|
}, |
|
{ |
|
"epoch": 1.81, |
|
"eval_hfsummary_accuracy": 0.6713720037481486, |
|
"eval_hfsummary_loss": 0.7460022568702698, |
|
"eval_hfsummary_runtime": 1243.8848, |
|
"eval_hfsummary_samples_per_second": 26.597, |
|
"eval_hfsummary_steps_per_second": 2.66, |
|
"step": 3800 |
|
}, |
|
{ |
|
"epoch": 1.81, |
|
"eval_gptsynthetic_accuracy": 0.9987933634992459, |
|
"eval_gptsynthetic_loss": 0.005401162896305323, |
|
"eval_gptsynthetic_runtime": 46.0774, |
|
"eval_gptsynthetic_samples_per_second": 71.944, |
|
"eval_gptsynthetic_steps_per_second": 7.205, |
|
"step": 3800 |
|
}, |
|
{ |
|
"epoch": 1.82, |
|
"learning_rate": 6.250366702709971e-06, |
|
"loss": 0.3678, |
|
"step": 3810 |
|
}, |
|
{ |
|
"epoch": 1.82, |
|
"learning_rate": 6.231067800657502e-06, |
|
"loss": 0.3446, |
|
"step": 3820 |
|
}, |
|
{ |
|
"epoch": 1.83, |
|
"learning_rate": 6.211749354859871e-06, |
|
"loss": 0.3247, |
|
"step": 3830 |
|
}, |
|
{ |
|
"epoch": 1.83, |
|
"learning_rate": 6.192411672005937e-06, |
|
"loss": 0.3639, |
|
"step": 3840 |
|
}, |
|
{ |
|
"epoch": 1.84, |
|
"learning_rate": 6.173055059089958e-06, |
|
"loss": 0.3581, |
|
"step": 3850 |
|
}, |
|
{ |
|
"epoch": 1.84, |
|
"learning_rate": 6.1536798234067175e-06, |
|
"loss": 0.3423, |
|
"step": 3860 |
|
}, |
|
{ |
|
"epoch": 1.85, |
|
"learning_rate": 6.134286272546639e-06, |
|
"loss": 0.3623, |
|
"step": 3870 |
|
}, |
|
{ |
|
"epoch": 1.85, |
|
"learning_rate": 6.11487471439091e-06, |
|
"loss": 0.321, |
|
"step": 3880 |
|
}, |
|
{ |
|
"epoch": 1.86, |
|
"learning_rate": 6.0954454571065925e-06, |
|
"loss": 0.3363, |
|
"step": 3890 |
|
}, |
|
{ |
|
"epoch": 1.86, |
|
"learning_rate": 6.0759988091417266e-06, |
|
"loss": 0.3392, |
|
"step": 3900 |
|
}, |
|
{ |
|
"epoch": 1.87, |
|
"learning_rate": 6.0565350792204406e-06, |
|
"loss": 0.3458, |
|
"step": 3910 |
|
}, |
|
{ |
|
"epoch": 1.87, |
|
"learning_rate": 6.037054576338044e-06, |
|
"loss": 0.3742, |
|
"step": 3920 |
|
}, |
|
{ |
|
"epoch": 1.87, |
|
"learning_rate": 6.0175576097561265e-06, |
|
"loss": 0.3518, |
|
"step": 3930 |
|
}, |
|
{ |
|
"epoch": 1.88, |
|
"learning_rate": 5.998044488997645e-06, |
|
"loss": 0.3489, |
|
"step": 3940 |
|
}, |
|
{ |
|
"epoch": 1.88, |
|
"learning_rate": 5.9785155238420135e-06, |
|
"loss": 0.3411, |
|
"step": 3950 |
|
}, |
|
{ |
|
"epoch": 1.89, |
|
"learning_rate": 5.958971024320179e-06, |
|
"loss": 0.3326, |
|
"step": 3960 |
|
}, |
|
{ |
|
"epoch": 1.89, |
|
"learning_rate": 5.9394113007097045e-06, |
|
"loss": 0.322, |
|
"step": 3970 |
|
}, |
|
{ |
|
"epoch": 1.9, |
|
"learning_rate": 5.919836663529845e-06, |
|
"loss": 0.3343, |
|
"step": 3980 |
|
}, |
|
{ |
|
"epoch": 1.9, |
|
"learning_rate": 5.900247423536611e-06, |
|
"loss": 0.3551, |
|
"step": 3990 |
|
}, |
|
{ |
|
"epoch": 1.91, |
|
"learning_rate": 5.880643891717842e-06, |
|
"loss": 0.3344, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 1.91, |
|
"eval_webgpt_accuracy": 0.5906537282941777, |
|
"eval_webgpt_loss": 0.6790869832038879, |
|
"eval_webgpt_runtime": 63.367, |
|
"eval_webgpt_samples_per_second": 61.799, |
|
"eval_webgpt_steps_per_second": 6.186, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 1.91, |
|
"eval_hfsummary_accuracy": 0.6683795302723453, |
|
"eval_hfsummary_loss": 0.7165785431861877, |
|
"eval_hfsummary_runtime": 1243.3678, |
|
"eval_hfsummary_samples_per_second": 26.608, |
|
"eval_hfsummary_steps_per_second": 2.661, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 1.91, |
|
"eval_gptsynthetic_accuracy": 0.9984917043740573, |
|
"eval_gptsynthetic_loss": 0.005581920500844717, |
|
"eval_gptsynthetic_runtime": 46.0581, |
|
"eval_gptsynthetic_samples_per_second": 71.974, |
|
"eval_gptsynthetic_steps_per_second": 7.208, |
|
"step": 4000 |
|
} |
|
], |
|
"max_steps": 4192, |
|
"num_train_epochs": 2, |
|
"total_flos": 2.4953197931002634e+17, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|