| { | |
| "best_metric": 0.7640588283538818, | |
| "best_model_checkpoint": "./model_20230703_attempt1/checkpoint-400", | |
| "epoch": 13.445378151260504, | |
| "global_step": 1200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 2.9999999999999997e-05, | |
| "loss": 2.5488, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 5.9999999999999995e-05, | |
| "loss": 2.3131, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 8.999999999999999e-05, | |
| "loss": 2.1798, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00011999999999999999, | |
| "loss": 1.6567, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.000147, | |
| "loss": 1.2086, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.00017699999999999997, | |
| "loss": 0.8952, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.00020699999999999996, | |
| "loss": 1.0646, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.000237, | |
| "loss": 0.7725, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 0.000267, | |
| "loss": 0.8192, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.00029699999999999996, | |
| "loss": 0.7403, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.00029764397905759156, | |
| "loss": 0.6473, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.0002950261780104712, | |
| "loss": 0.7524, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.00029240837696335075, | |
| "loss": 0.6775, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 0.00028979057591623034, | |
| "loss": 0.5884, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.00028717277486910994, | |
| "loss": 0.7117, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 0.00028455497382198953, | |
| "loss": 0.5925, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 0.00028193717277486907, | |
| "loss": 0.6675, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 0.00027931937172774867, | |
| "loss": 0.6126, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 0.00027670157068062826, | |
| "loss": 0.557, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 0.00027408376963350786, | |
| "loss": 0.5362, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "eval_loss": 0.8287817239761353, | |
| "eval_runtime": 34.4542, | |
| "eval_samples_per_second": 11.61, | |
| "eval_steps_per_second": 1.451, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 0.0002714659685863874, | |
| "loss": 0.4951, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 0.000268848167539267, | |
| "loss": 0.4904, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 0.0002662303664921466, | |
| "loss": 0.4544, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 0.0002636125654450262, | |
| "loss": 0.5705, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 0.0002609947643979057, | |
| "loss": 0.5654, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 0.0002583769633507853, | |
| "loss": 0.445, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 0.0002557591623036649, | |
| "loss": 0.3859, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 0.0002531413612565445, | |
| "loss": 0.5015, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 0.00025052356020942404, | |
| "loss": 0.4177, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 0.00024790575916230364, | |
| "loss": 0.4322, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 0.00024528795811518323, | |
| "loss": 0.3028, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 0.00024267015706806283, | |
| "loss": 0.3839, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 0.0002400523560209424, | |
| "loss": 0.3121, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 0.00023743455497382196, | |
| "loss": 0.5043, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 0.00023481675392670153, | |
| "loss": 0.3569, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 0.00023219895287958115, | |
| "loss": 0.2791, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "learning_rate": 0.00022958115183246072, | |
| "loss": 0.2648, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 0.00022696335078534028, | |
| "loss": 0.2998, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 0.00022434554973821985, | |
| "loss": 0.3416, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 0.00022172774869109947, | |
| "loss": 0.3044, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "eval_loss": 0.7640588283538818, | |
| "eval_runtime": 34.4837, | |
| "eval_samples_per_second": 11.6, | |
| "eval_steps_per_second": 1.45, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 0.00021910994764397904, | |
| "loss": 0.4006, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 4.71, | |
| "learning_rate": 0.0002164921465968586, | |
| "loss": 0.2957, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "learning_rate": 0.0002138743455497382, | |
| "loss": 0.3241, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "learning_rate": 0.0002112565445026178, | |
| "loss": 0.2763, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 0.00020863874345549736, | |
| "loss": 0.2206, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 5.15, | |
| "learning_rate": 0.00020602094240837693, | |
| "loss": 0.2224, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 5.27, | |
| "learning_rate": 0.00020340314136125653, | |
| "loss": 0.1844, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 0.00020078534031413612, | |
| "loss": 0.2651, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "learning_rate": 0.0001981675392670157, | |
| "loss": 0.2708, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 0.00019554973821989528, | |
| "loss": 0.2604, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.71, | |
| "learning_rate": 0.00019293193717277485, | |
| "loss": 0.2326, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 5.83, | |
| "learning_rate": 0.00019031413612565444, | |
| "loss": 0.2328, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "learning_rate": 0.000187696335078534, | |
| "loss": 0.2052, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 0.0001850785340314136, | |
| "loss": 0.3427, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 6.16, | |
| "learning_rate": 0.00018246073298429317, | |
| "loss": 0.164, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 6.27, | |
| "learning_rate": 0.00017984293193717277, | |
| "loss": 0.172, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 6.39, | |
| "learning_rate": 0.00017722513089005233, | |
| "loss": 0.2155, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "learning_rate": 0.00017460732984293193, | |
| "loss": 0.1079, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 6.61, | |
| "learning_rate": 0.0001719895287958115, | |
| "loss": 0.1656, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "learning_rate": 0.0001693717277486911, | |
| "loss": 0.1915, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "eval_loss": 0.8178408145904541, | |
| "eval_runtime": 34.3695, | |
| "eval_samples_per_second": 11.638, | |
| "eval_steps_per_second": 1.455, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 0.00016675392670157069, | |
| "loss": 0.2569, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 6.95, | |
| "learning_rate": 0.00016413612565445025, | |
| "loss": 0.2147, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 0.00016151832460732982, | |
| "loss": 0.2239, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 7.17, | |
| "learning_rate": 0.00015890052356020941, | |
| "loss": 0.1435, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 7.28, | |
| "learning_rate": 0.000156282722513089, | |
| "loss": 0.201, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 7.39, | |
| "learning_rate": 0.00015366492146596858, | |
| "loss": 0.1336, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 7.51, | |
| "learning_rate": 0.00015104712041884814, | |
| "loss": 0.1271, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 7.62, | |
| "learning_rate": 0.00014842931937172774, | |
| "loss": 0.1659, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 7.73, | |
| "learning_rate": 0.00014581151832460733, | |
| "loss": 0.141, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 7.84, | |
| "learning_rate": 0.0001431937172774869, | |
| "loss": 0.1463, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.96, | |
| "learning_rate": 0.0001405759162303665, | |
| "loss": 0.1573, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 0.00013795811518324606, | |
| "loss": 0.1132, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 8.18, | |
| "learning_rate": 0.00013534031413612566, | |
| "loss": 0.0951, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "learning_rate": 0.00013272251308900522, | |
| "loss": 0.1343, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "learning_rate": 0.00013010471204188482, | |
| "loss": 0.1275, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 8.52, | |
| "learning_rate": 0.00012748691099476439, | |
| "loss": 0.1846, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 8.63, | |
| "learning_rate": 0.00012486910994764398, | |
| "loss": 0.1475, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 8.74, | |
| "learning_rate": 0.00012225130890052355, | |
| "loss": 0.1069, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 8.85, | |
| "learning_rate": 0.00011963350785340314, | |
| "loss": 0.1219, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "learning_rate": 0.00011701570680628271, | |
| "loss": 0.1004, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "eval_loss": 0.917429506778717, | |
| "eval_runtime": 34.3427, | |
| "eval_samples_per_second": 11.647, | |
| "eval_steps_per_second": 1.456, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "learning_rate": 0.0001143979057591623, | |
| "loss": 0.1, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 9.19, | |
| "learning_rate": 0.00011178010471204187, | |
| "loss": 0.0793, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "learning_rate": 0.00010916230366492146, | |
| "loss": 0.1194, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "learning_rate": 0.00010654450261780103, | |
| "loss": 0.0992, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 0.00010392670157068063, | |
| "loss": 0.1368, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 9.64, | |
| "learning_rate": 0.0001013089005235602, | |
| "loss": 0.0942, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 9.75, | |
| "learning_rate": 9.869109947643979e-05, | |
| "loss": 0.0907, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 9.86, | |
| "learning_rate": 9.607329842931936e-05, | |
| "loss": 0.1154, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 9.97, | |
| "learning_rate": 9.345549738219895e-05, | |
| "loss": 0.1075, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 10.08, | |
| "learning_rate": 9.083769633507853e-05, | |
| "loss": 0.0699, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 10.2, | |
| "learning_rate": 8.821989528795811e-05, | |
| "loss": 0.0785, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 10.31, | |
| "learning_rate": 8.560209424083769e-05, | |
| "loss": 0.0849, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 10.42, | |
| "learning_rate": 8.298429319371727e-05, | |
| "loss": 0.1372, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 10.53, | |
| "learning_rate": 8.036649214659685e-05, | |
| "loss": 0.1006, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 10.64, | |
| "learning_rate": 7.774869109947644e-05, | |
| "loss": 0.0792, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 10.76, | |
| "learning_rate": 7.513089005235602e-05, | |
| "loss": 0.0723, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 10.87, | |
| "learning_rate": 7.25130890052356e-05, | |
| "loss": 0.071, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 10.98, | |
| "learning_rate": 6.989528795811518e-05, | |
| "loss": 0.0798, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 11.09, | |
| "learning_rate": 6.727748691099476e-05, | |
| "loss": 0.0936, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "learning_rate": 6.465968586387434e-05, | |
| "loss": 0.0676, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "eval_loss": 1.1027586460113525, | |
| "eval_runtime": 34.3596, | |
| "eval_samples_per_second": 11.642, | |
| "eval_steps_per_second": 1.455, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 11.32, | |
| "learning_rate": 6.204188481675392e-05, | |
| "loss": 0.0702, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 11.43, | |
| "learning_rate": 5.94240837696335e-05, | |
| "loss": 0.0761, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 11.54, | |
| "learning_rate": 5.680628272251308e-05, | |
| "loss": 0.0727, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 11.65, | |
| "learning_rate": 5.4188481675392663e-05, | |
| "loss": 0.0688, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 11.76, | |
| "learning_rate": 5.1570680628272244e-05, | |
| "loss": 0.0781, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 11.88, | |
| "learning_rate": 4.8952879581151825e-05, | |
| "loss": 0.068, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 11.99, | |
| "learning_rate": 4.6335078534031406e-05, | |
| "loss": 0.0682, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 12.1, | |
| "learning_rate": 4.371727748691099e-05, | |
| "loss": 0.0591, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 12.21, | |
| "learning_rate": 4.109947643979057e-05, | |
| "loss": 0.0579, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 12.32, | |
| "learning_rate": 3.8481675392670156e-05, | |
| "loss": 0.0632, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 12.44, | |
| "learning_rate": 3.5863874345549736e-05, | |
| "loss": 0.0635, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 12.55, | |
| "learning_rate": 3.324607329842932e-05, | |
| "loss": 0.0857, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 12.66, | |
| "learning_rate": 3.06282722513089e-05, | |
| "loss": 0.0656, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 12.77, | |
| "learning_rate": 2.801047120418848e-05, | |
| "loss": 0.0649, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 12.89, | |
| "learning_rate": 2.539267015706806e-05, | |
| "loss": 0.0572, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 2.277486910994764e-05, | |
| "loss": 0.0638, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 13.11, | |
| "learning_rate": 2.0157068062827222e-05, | |
| "loss": 0.0556, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 13.22, | |
| "learning_rate": 1.7539267015706803e-05, | |
| "loss": 0.0529, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 13.33, | |
| "learning_rate": 1.4921465968586385e-05, | |
| "loss": 0.0513, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 13.45, | |
| "learning_rate": 1.2303664921465966e-05, | |
| "loss": 0.0845, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 13.45, | |
| "eval_loss": 1.133836030960083, | |
| "eval_runtime": 34.4155, | |
| "eval_samples_per_second": 11.623, | |
| "eval_steps_per_second": 1.453, | |
| "step": 1200 | |
| } | |
| ], | |
| "max_steps": 1246, | |
| "num_train_epochs": 14, | |
| "total_flos": 3.984965106597888e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |