| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.5259612685938815, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 8e-08, | |
| "loss": 11.8368, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 1.6e-07, | |
| "loss": 11.103, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 2.4e-07, | |
| "loss": 11.6842, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 3.2e-07, | |
| "loss": 11.7279, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 4e-07, | |
| "loss": 11.6439, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.8e-07, | |
| "loss": 11.1668, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 5.6e-07, | |
| "loss": 11.5542, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 6.4e-07, | |
| "loss": 11.2201, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 7.2e-07, | |
| "loss": 10.0118, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 8e-07, | |
| "loss": 9.2907, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 8.799999999999999e-07, | |
| "loss": 8.9216, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 9.6e-07, | |
| "loss": 8.9268, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 1.04e-06, | |
| "loss": 8.6545, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 1.12e-06, | |
| "loss": 8.7888, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 1.2e-06, | |
| "loss": 9.0201, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.28e-06, | |
| "loss": 8.0512, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 1.3600000000000001e-06, | |
| "loss": 5.9266, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 1.44e-06, | |
| "loss": 5.3527, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 1.5199999999999998e-06, | |
| "loss": 4.9847, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 1.6e-06, | |
| "loss": 4.8117, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_webgpt_accuracy": 0.3513391153050626, | |
| "eval_webgpt_loss": 4.794132709503174, | |
| "eval_webgpt_runtime": 39.5471, | |
| "eval_webgpt_samples_per_second": 90.095, | |
| "eval_webgpt_steps_per_second": 22.53, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 1.6799999999999998e-06, | |
| "loss": 4.7854, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 1.7599999999999999e-06, | |
| "loss": 4.5555, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 1.84e-06, | |
| "loss": 4.4819, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 1.92e-06, | |
| "loss": 4.3504, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 2e-06, | |
| "loss": 4.1471, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 2.08e-06, | |
| "loss": 4.4132, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 2.16e-06, | |
| "loss": 4.1309, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 2.24e-06, | |
| "loss": 4.4153, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 2.32e-06, | |
| "loss": 4.3256, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 2.4e-06, | |
| "loss": 4.3798, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 2.48e-06, | |
| "loss": 4.3665, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 2.56e-06, | |
| "loss": 4.2545, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 2.64e-06, | |
| "loss": 4.1912, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 2.7200000000000002e-06, | |
| "loss": 4.2359, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 2.8e-06, | |
| "loss": 4.2145, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 2.88e-06, | |
| "loss": 4.0716, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 2.96e-06, | |
| "loss": 4.0429, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 3.0399999999999997e-06, | |
| "loss": 4.1404, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 3.1199999999999998e-06, | |
| "loss": 3.9878, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 3.2e-06, | |
| "loss": 4.1169, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_webgpt_accuracy": 0.3585799846229516, | |
| "eval_webgpt_loss": 3.982567310333252, | |
| "eval_webgpt_runtime": 39.5324, | |
| "eval_webgpt_samples_per_second": 90.129, | |
| "eval_webgpt_steps_per_second": 22.538, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 3.2799999999999995e-06, | |
| "loss": 3.9303, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 3.3599999999999996e-06, | |
| "loss": 4.0221, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 3.4399999999999997e-06, | |
| "loss": 3.8958, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 3.5199999999999998e-06, | |
| "loss": 3.8434, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 3.6e-06, | |
| "loss": 4.0083, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 3.68e-06, | |
| "loss": 3.9172, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 3.7599999999999996e-06, | |
| "loss": 4.0955, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 3.84e-06, | |
| "loss": 4.1085, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 3.92e-06, | |
| "loss": 4.149, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 4e-06, | |
| "loss": 4.0345, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 4.08e-06, | |
| "loss": 3.9929, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 4.16e-06, | |
| "loss": 3.7969, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 4.24e-06, | |
| "loss": 3.6102, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 4.32e-06, | |
| "loss": 3.744, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 4.4e-06, | |
| "loss": 3.8242, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 4.48e-06, | |
| "loss": 3.8092, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 4.5599999999999995e-06, | |
| "loss": 3.8835, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 4.64e-06, | |
| "loss": 3.7335, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.72e-06, | |
| "loss": 3.8462, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 4.8e-06, | |
| "loss": 3.7422, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_webgpt_accuracy": 0.36334147099294245, | |
| "eval_webgpt_loss": 3.8047080039978027, | |
| "eval_webgpt_runtime": 39.6534, | |
| "eval_webgpt_samples_per_second": 89.854, | |
| "eval_webgpt_steps_per_second": 22.47, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 4.88e-06, | |
| "loss": 3.7068, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 4.96e-06, | |
| "loss": 3.69, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 5.04e-06, | |
| "loss": 3.65, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 5.12e-06, | |
| "loss": 3.6972, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 5.2e-06, | |
| "loss": 3.7907, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 5.28e-06, | |
| "loss": 3.7505, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 5.36e-06, | |
| "loss": 3.8109, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 5.4400000000000004e-06, | |
| "loss": 3.8612, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 5.52e-06, | |
| "loss": 3.7077, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 5.6e-06, | |
| "loss": 3.8546, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 5.68e-06, | |
| "loss": 3.8801, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 5.76e-06, | |
| "loss": 3.7939, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 5.84e-06, | |
| "loss": 3.6499, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 5.92e-06, | |
| "loss": 4.0742, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 6e-06, | |
| "loss": 3.8072, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 6.079999999999999e-06, | |
| "loss": 3.7959, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 6.1599999999999995e-06, | |
| "loss": 3.7556, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 6.2399999999999995e-06, | |
| "loss": 3.8263, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 6.32e-06, | |
| "loss": 3.9162, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 6.4e-06, | |
| "loss": 3.7188, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "eval_webgpt_accuracy": 0.3654184548784667, | |
| "eval_webgpt_loss": 3.7724239826202393, | |
| "eval_webgpt_runtime": 39.717, | |
| "eval_webgpt_samples_per_second": 89.71, | |
| "eval_webgpt_steps_per_second": 22.434, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 6.48e-06, | |
| "loss": 3.7938, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 6.559999999999999e-06, | |
| "loss": 3.6401, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 6.639999999999999e-06, | |
| "loss": 3.7234, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 6.719999999999999e-06, | |
| "loss": 3.6234, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 6.799999999999999e-06, | |
| "loss": 3.7614, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 6.879999999999999e-06, | |
| "loss": 3.7001, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 6.9599999999999994e-06, | |
| "loss": 3.8073, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 7.0399999999999995e-06, | |
| "loss": 3.6328, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 7.12e-06, | |
| "loss": 3.9974, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 7.2e-06, | |
| "loss": 3.6881, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 7.28e-06, | |
| "loss": 3.7669, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 7.36e-06, | |
| "loss": 3.5966, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 7.44e-06, | |
| "loss": 3.748, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 7.519999999999999e-06, | |
| "loss": 3.7903, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 7.599999999999999e-06, | |
| "loss": 3.7352, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 7.68e-06, | |
| "loss": 3.5784, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 7.76e-06, | |
| "loss": 3.5768, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 7.84e-06, | |
| "loss": 3.7704, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 7.92e-06, | |
| "loss": 3.6493, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 8e-06, | |
| "loss": 3.6712, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "eval_webgpt_accuracy": 0.36571671085272284, | |
| "eval_webgpt_loss": 3.766130208969116, | |
| "eval_webgpt_runtime": 39.6543, | |
| "eval_webgpt_samples_per_second": 89.852, | |
| "eval_webgpt_steps_per_second": 22.469, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 7.941733430444282e-06, | |
| "loss": 3.8326, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 7.883466860888565e-06, | |
| "loss": 3.6731, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 7.825200291332846e-06, | |
| "loss": 3.6476, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 7.76693372177713e-06, | |
| "loss": 3.7165, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 7.708667152221412e-06, | |
| "loss": 3.6997, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 7.650400582665694e-06, | |
| "loss": 3.7811, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 7.592134013109977e-06, | |
| "loss": 3.6025, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 7.533867443554261e-06, | |
| "loss": 3.8738, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 7.475600873998543e-06, | |
| "loss": 3.7493, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 7.4173343044428255e-06, | |
| "loss": 3.8542, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 7.359067734887107e-06, | |
| "loss": 3.4613, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 7.300801165331391e-06, | |
| "loss": 3.7582, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 7.242534595775673e-06, | |
| "loss": 3.6542, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 7.184268026219956e-06, | |
| "loss": 3.6867, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 7.126001456664238e-06, | |
| "loss": 3.6046, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 7.067734887108521e-06, | |
| "loss": 3.7645, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 7.009468317552803e-06, | |
| "loss": 3.8135, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 6.951201747997087e-06, | |
| "loss": 3.8123, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 6.8929351784413686e-06, | |
| "loss": 3.743, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 6.834668608885651e-06, | |
| "loss": 3.8658, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "eval_webgpt_accuracy": 0.36873122296639843, | |
| "eval_webgpt_loss": 3.7487637996673584, | |
| "eval_webgpt_runtime": 39.5934, | |
| "eval_webgpt_samples_per_second": 89.99, | |
| "eval_webgpt_steps_per_second": 22.504, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 6.776402039329934e-06, | |
| "loss": 3.7027, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 6.718135469774217e-06, | |
| "loss": 3.6325, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 6.659868900218499e-06, | |
| "loss": 3.7518, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 6.601602330662782e-06, | |
| "loss": 3.5668, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 6.543335761107064e-06, | |
| "loss": 3.7264, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 6.485069191551347e-06, | |
| "loss": 3.5763, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 6.426802621995629e-06, | |
| "loss": 3.7075, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 6.3685360524399125e-06, | |
| "loss": 3.6597, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 6.310269482884194e-06, | |
| "loss": 3.8605, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 6.252002913328477e-06, | |
| "loss": 3.7568, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 6.19373634377276e-06, | |
| "loss": 3.6982, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 6.135469774217043e-06, | |
| "loss": 3.8453, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 6.0772032046613246e-06, | |
| "loss": 3.8372, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 6.018936635105608e-06, | |
| "loss": 3.5192, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 5.96067006554989e-06, | |
| "loss": 3.7037, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 5.902403495994173e-06, | |
| "loss": 3.4107, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 5.844136926438456e-06, | |
| "loss": 3.5911, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 5.785870356882738e-06, | |
| "loss": 3.7885, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 5.72760378732702e-06, | |
| "loss": 3.5698, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 5.669337217771304e-06, | |
| "loss": 3.6686, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "eval_webgpt_accuracy": 0.3689563590269814, | |
| "eval_webgpt_loss": 3.7346436977386475, | |
| "eval_webgpt_runtime": 39.5649, | |
| "eval_webgpt_samples_per_second": 90.055, | |
| "eval_webgpt_steps_per_second": 22.52, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 5.611070648215586e-06, | |
| "loss": 3.7345, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 5.5528040786598685e-06, | |
| "loss": 3.8317, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 5.49453750910415e-06, | |
| "loss": 3.7166, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 5.436270939548434e-06, | |
| "loss": 3.7229, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 5.378004369992716e-06, | |
| "loss": 3.4168, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 5.319737800436999e-06, | |
| "loss": 3.9726, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 5.2614712308812814e-06, | |
| "loss": 3.8361, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 5.203204661325564e-06, | |
| "loss": 3.5802, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 5.144938091769846e-06, | |
| "loss": 3.9139, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 5.08667152221413e-06, | |
| "loss": 3.5712, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 5.028404952658412e-06, | |
| "loss": 3.6163, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 4.970138383102694e-06, | |
| "loss": 3.5838, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 4.911871813546977e-06, | |
| "loss": 3.8421, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 4.85360524399126e-06, | |
| "loss": 3.6685, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 4.795338674435542e-06, | |
| "loss": 3.6735, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 4.737072104879825e-06, | |
| "loss": 3.7081, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 4.678805535324107e-06, | |
| "loss": 3.7911, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.62053896576839e-06, | |
| "loss": 3.7296, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 4.562272396212672e-06, | |
| "loss": 3.6247, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 4.5040058266569555e-06, | |
| "loss": 3.5536, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "eval_webgpt_accuracy": 0.3708860692093479, | |
| "eval_webgpt_loss": 3.709535837173462, | |
| "eval_webgpt_runtime": 39.7168, | |
| "eval_webgpt_samples_per_second": 89.71, | |
| "eval_webgpt_steps_per_second": 22.434, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 4.4457392571012374e-06, | |
| "loss": 3.3749, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 4.38747268754552e-06, | |
| "loss": 3.6204, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 4.329206117989803e-06, | |
| "loss": 3.6534, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 4.270939548434086e-06, | |
| "loss": 3.769, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 4.212672978878368e-06, | |
| "loss": 3.4946, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 4.154406409322651e-06, | |
| "loss": 3.4767, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 4.096139839766933e-06, | |
| "loss": 4.0031, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 4.037873270211216e-06, | |
| "loss": 3.5801, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 3.979606700655499e-06, | |
| "loss": 3.7839, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 3.921340131099781e-06, | |
| "loss": 3.389, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 3.863073561544064e-06, | |
| "loss": 3.5539, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 3.8048069919883465e-06, | |
| "loss": 3.7058, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 3.746540422432629e-06, | |
| "loss": 3.6334, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 3.6882738528769115e-06, | |
| "loss": 3.5749, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 3.6300072833211943e-06, | |
| "loss": 3.5804, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "learning_rate": 3.571740713765477e-06, | |
| "loss": 3.5608, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 3.5134741442097594e-06, | |
| "loss": 3.8894, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 3.455207574654042e-06, | |
| "loss": 3.4811, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 3.396941005098325e-06, | |
| "loss": 3.5773, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 3.338674435542607e-06, | |
| "loss": 3.4083, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "eval_webgpt_accuracy": 0.37110112351569674, | |
| "eval_webgpt_loss": 3.7004425525665283, | |
| "eval_webgpt_runtime": 39.7103, | |
| "eval_webgpt_samples_per_second": 89.725, | |
| "eval_webgpt_steps_per_second": 22.438, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 3.28040786598689e-06, | |
| "loss": 3.5236, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 3.2221412964311723e-06, | |
| "loss": 3.6029, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 3.163874726875455e-06, | |
| "loss": 3.6437, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "learning_rate": 3.105608157319738e-06, | |
| "loss": 3.5334, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 3.04734158776402e-06, | |
| "loss": 3.5593, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 2.989075018208303e-06, | |
| "loss": 3.8394, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 2.9308084486525856e-06, | |
| "loss": 3.5156, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 2.872541879096868e-06, | |
| "loss": 3.5369, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 2.8142753095411507e-06, | |
| "loss": 3.5477, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 2.7560087399854335e-06, | |
| "loss": 3.6827, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 2.697742170429716e-06, | |
| "loss": 3.5929, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 2.6394756008739986e-06, | |
| "loss": 3.7087, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 2.581209031318281e-06, | |
| "loss": 3.4699, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 2.5229424617625636e-06, | |
| "loss": 3.5238, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 2.4646758922068464e-06, | |
| "loss": 3.6169, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 2.4064093226511287e-06, | |
| "loss": 3.536, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 2.3481427530954115e-06, | |
| "loss": 3.4735, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 2.2898761835396942e-06, | |
| "loss": 3.616, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 2.2316096139839766e-06, | |
| "loss": 3.4397, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 2.1733430444282593e-06, | |
| "loss": 3.6732, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "eval_webgpt_accuracy": 0.37167770768266506, | |
| "eval_webgpt_loss": 3.6911776065826416, | |
| "eval_webgpt_runtime": 39.7365, | |
| "eval_webgpt_samples_per_second": 89.666, | |
| "eval_webgpt_steps_per_second": 22.423, | |
| "step": 2000 | |
| } | |
| ], | |
| "max_steps": 2373, | |
| "num_train_epochs": 3, | |
| "total_flos": 2456896174817280.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |