| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.9926873857404024, | |
| "eval_steps": 500, | |
| "global_step": 1092, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 4.545454545454545e-08, | |
| "loss": 6.3231, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 9.09090909090909e-08, | |
| "loss": 5.5694, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1.3636363636363635e-07, | |
| "loss": 3.253, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 1.818181818181818e-07, | |
| "loss": 1.3696, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 2.2727272727272726e-07, | |
| "loss": 0.5924, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 2.727272727272727e-07, | |
| "loss": 0.2712, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 3.1818181818181815e-07, | |
| "loss": 0.1977, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 3.636363636363636e-07, | |
| "loss": 0.1782, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.090909090909091e-07, | |
| "loss": 0.1586, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 4.545454545454545e-07, | |
| "loss": 0.1282, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 5e-07, | |
| "loss": 0.083, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 4.998720766742264e-07, | |
| "loss": 0.0686, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 4.994884376119235e-07, | |
| "loss": 0.0636, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 4.988494754241696e-07, | |
| "loss": 0.0618, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 4.979558440163095e-07, | |
| "loss": 0.0609, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 4.968084579187565e-07, | |
| "loss": 0.0572, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 4.954084913510755e-07, | |
| "loss": 0.0576, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 4.937573770203005e-07, | |
| "loss": 0.0616, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 4.918568046547231e-07, | |
| "loss": 0.058, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 4.897087192746464e-07, | |
| "loss": 0.0581, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 4.873153192018773e-07, | |
| "loss": 0.0561, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.846790538099934e-07, | |
| "loss": 0.0527, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 4.818026210176872e-07, | |
| "loss": 0.0527, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 4.786889645277519e-07, | |
| "loss": 0.0546, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 4.7534127081453554e-07, | |
| "loss": 0.0513, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 4.71762965862946e-07, | |
| "loss": 0.0489, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 4.679577116623435e-07, | |
| "loss": 0.0511, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.051615118980407715, | |
| "eval_runtime": 48.6349, | |
| "eval_samples_per_second": 7.135, | |
| "eval_steps_per_second": 0.596, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 4.639294024589101e-07, | |
| "loss": 0.0476, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 4.596821607703302e-07, | |
| "loss": 0.0486, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 4.5522033316686106e-07, | |
| "loss": 0.0473, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 4.5054848582311134e-07, | |
| "loss": 0.0501, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 4.456713998450786e-07, | |
| "loss": 0.0435, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 4.405940663772302e-07, | |
| "loss": 0.0393, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 4.353216814946321e-07, | |
| "loss": 0.0489, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 4.2985964088535603e-07, | |
| "loss": 0.039, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 4.2421353432860386e-07, | |
| "loss": 0.045, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 4.1838913997420285e-07, | |
| "loss": 0.0389, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 4.1239241842932446e-07, | |
| "loss": 0.0407, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 4.0622950665847893e-07, | |
| "loss": 0.0442, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 3.9990671170302747e-07, | |
| "loss": 0.04, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 3.934305042266413e-07, | |
| "loss": 0.0398, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 3.868075118933106e-07, | |
| "loss": 0.0478, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 3.8004451258468224e-07, | |
| "loss": 0.0426, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 3.7314842746366625e-07, | |
| "loss": 0.0398, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 3.6612631389141073e-07, | |
| "loss": 0.0407, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 3.5898535820489257e-07, | |
| "loss": 0.0418, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 3.5173286836251683e-07, | |
| "loss": 0.0411, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 3.4437626646524965e-07, | |
| "loss": 0.0363, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 3.369230811609397e-07, | |
| "loss": 0.0397, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 3.2938093993960107e-07, | |
| "loss": 0.0439, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 3.217575613275414e-07, | |
| "loss": 0.0437, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 3.1406074698832646e-07, | |
| "loss": 0.0445, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 3.0629837373866113e-07, | |
| "loss": 0.0401, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 2.984783854873614e-07, | |
| "loss": 0.0412, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.04081326350569725, | |
| "eval_runtime": 48.5536, | |
| "eval_samples_per_second": 7.147, | |
| "eval_steps_per_second": 0.597, | |
| "step": 547 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 2.9060878510566426e-07, | |
| "loss": 0.0354, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 2.8269762623719636e-07, | |
| "loss": 0.0325, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 2.747530050559831e-07, | |
| "loss": 0.0327, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 2.6678305198093267e-07, | |
| "loss": 0.0373, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 2.587959233552739e-07, | |
| "loss": 0.0377, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 2.507997930994643e-07, | |
| "loss": 0.0328, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 2.42802844346109e-07, | |
| "loss": 0.0357, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 2.3481326106545248e-07, | |
| "loss": 0.0334, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 2.2683921969001343e-07, | |
| "loss": 0.0318, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 2.1888888074693235e-07, | |
| "loss": 0.0324, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 2.109703805065985e-07, | |
| "loss": 0.0321, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 2.0309182265609888e-07, | |
| "loss": 0.0328, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 1.952612700060137e-07, | |
| "loss": 0.0332, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 1.874867362390442e-07, | |
| "loss": 0.0329, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 1.7977617770891678e-07, | |
| "loss": 0.0328, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 1.7213748529795756e-07, | |
| "loss": 0.0272, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 1.645784763416686e-07, | |
| "loss": 0.0373, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 1.5710688662857186e-07, | |
| "loss": 0.031, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 1.4973036248350665e-07, | |
| "loss": 0.0342, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 1.4245645294248333e-07, | |
| "loss": 0.037, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 1.352926020271006e-07, | |
| "loss": 0.0334, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 1.2824614112643428e-07, | |
| "loss": 0.034, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 1.2132428149419105e-07, | |
| "loss": 0.036, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 1.1453410686880843e-07, | |
| "loss": 0.0383, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 1.0788256622405165e-07, | |
| "loss": 0.0266, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 1.0137646665752717e-07, | |
| "loss": 0.0318, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 9.502246642438996e-08, | |
| "loss": 0.0336, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 8.882706812337432e-08, | |
| "loss": 0.032, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.03941405192017555, | |
| "eval_runtime": 48.6039, | |
| "eval_samples_per_second": 7.139, | |
| "eval_steps_per_second": 0.597, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 8.279661204212143e-08, | |
| "loss": 0.0254, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 7.693726966861405e-08, | |
| "loss": 0.0328, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 7.125503737535843e-08, | |
| "loss": 0.0231, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 6.575573028277659e-08, | |
| "loss": 0.0276, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 6.044497630809053e-08, | |
| "loss": 0.0239, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 5.5328210405786696e-08, | |
| "loss": 0.0251, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 5.041066900555646e-08, | |
| "loss": 0.025, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 3.29, | |
| "learning_rate": 4.569738465340414e-08, | |
| "loss": 0.0281, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 4.1193180861406965e-08, | |
| "loss": 0.0246, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 3.690266717139728e-08, | |
| "loss": 0.0252, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 3.2830234437619035e-08, | |
| "loss": 0.0293, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 2.8980050333186595e-08, | |
| "loss": 0.027, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 2.5356055084943763e-08, | |
| "loss": 0.025, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 2.1961957441088557e-08, | |
| "loss": 0.0255, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 1.8801230875689688e-08, | |
| "loss": 0.0273, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 1.5877110033980152e-08, | |
| "loss": 0.0291, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 1.3192587422064166e-08, | |
| "loss": 0.0273, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 1.0750410344426781e-08, | |
| "loss": 0.0303, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 8.553078092379123e-09, | |
| "loss": 0.0302, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 6.6028393863173935e-09, | |
| "loss": 0.024, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 4.901690074412185e-09, | |
| "loss": 0.0245, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 3.451371090084476e-09, | |
| "loss": 0.0211, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 2.2533666703574983e-09, | |
| "loss": 0.0224, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "learning_rate": 1.308902836908543e-09, | |
| "loss": 0.0201, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 6.189461413748098e-10, | |
| "loss": 0.0242, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 1.842026761973703e-10, | |
| "loss": 0.0301, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 5.117352015610033e-12, | |
| "loss": 0.0258, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "eval_loss": 0.04152332618832588, | |
| "eval_runtime": 48.6503, | |
| "eval_samples_per_second": 7.133, | |
| "eval_steps_per_second": 0.596, | |
| "step": 1092 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "step": 1092, | |
| "total_flos": 193702916259840.0, | |
| "train_loss": 0.20003320886702328, | |
| "train_runtime": 14516.5669, | |
| "train_samples_per_second": 1.806, | |
| "train_steps_per_second": 0.075 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1092, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "total_flos": 193702916259840.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |