| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 371760, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 3.125e-05, | |
| "loss": 6.2183, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 6.25e-05, | |
| "loss": 5.0127, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.375e-05, | |
| "loss": 4.6841, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.000125, | |
| "loss": 4.452, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.00015625, | |
| "loss": 4.2915, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.0001875, | |
| "loss": 4.181, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.00021875, | |
| "loss": 4.0744, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.00025, | |
| "loss": 3.9815, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.00028121875, | |
| "loss": 3.9114, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.0003124375, | |
| "loss": 3.8546, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00034368749999999997, | |
| "loss": 3.8053, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.00037490625, | |
| "loss": 3.7693, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.00040615625, | |
| "loss": 3.7387, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 0.00043740625, | |
| "loss": 3.6952, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.000468625, | |
| "loss": 3.6708, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.000499875, | |
| "loss": 3.6445, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 0.000531125, | |
| "loss": 3.6248, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.00056234375, | |
| "loss": 3.6071, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.3590285050576805, | |
| "eval_loss": 3.7804887294769287, | |
| "eval_runtime": 151.8602, | |
| "eval_samples_per_second": 381.397, | |
| "eval_steps_per_second": 5.959, | |
| "step": 18588 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.00059359375, | |
| "loss": 3.5742, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.0006248125, | |
| "loss": 3.5528, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 0.0006560625, | |
| "loss": 3.5368, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 0.0006873125, | |
| "loss": 3.5273, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.0007185000000000001, | |
| "loss": 3.5178, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.0007497500000000001, | |
| "loss": 3.5118, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.000781, | |
| "loss": 3.5063, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.00081225, | |
| "loss": 3.4865, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.00084346875, | |
| "loss": 3.4768, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 0.00087471875, | |
| "loss": 3.4753, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.00090596875, | |
| "loss": 3.46, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 0.00093721875, | |
| "loss": 3.4586, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.0009684375, | |
| "loss": 3.4509, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 0.0009996875, | |
| "loss": 3.4365, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 0.000997089121732988, | |
| "loss": 3.431, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 0.0009941458676712975, | |
| "loss": 3.4214, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 0.0009912026136096068, | |
| "loss": 3.408, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 0.0009882652460560396, | |
| "loss": 3.3984, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.000985321991994349, | |
| "loss": 3.3943, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.3805695233978648, | |
| "eval_loss": 3.579639434814453, | |
| "eval_runtime": 153.8621, | |
| "eval_samples_per_second": 376.434, | |
| "eval_steps_per_second": 5.882, | |
| "step": 37176 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.0009823816811867201, | |
| "loss": 3.3388, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 0.0009794384271250296, | |
| "loss": 3.3306, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 0.0009764951730633388, | |
| "loss": 3.3275, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 0.0009735519190016482, | |
| "loss": 3.3276, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 0.0009706086649399576, | |
| "loss": 3.3153, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 0.0009676712973863904, | |
| "loss": 3.3111, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 0.0009647280433246997, | |
| "loss": 3.3085, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 0.0009617847892630091, | |
| "loss": 3.3064, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 0.0009588444784553803, | |
| "loss": 3.3053, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 0.0009559012243936896, | |
| "loss": 3.2948, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 0.000952957970331999, | |
| "loss": 3.2911, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 0.0009500176595243702, | |
| "loss": 3.2898, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 0.0009470744054626795, | |
| "loss": 3.2839, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "learning_rate": 0.0009441311514009889, | |
| "loss": 3.2787, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 0.0009411908405933601, | |
| "loss": 3.2716, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 0.0009382475865316694, | |
| "loss": 3.2632, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 0.0009353043324699788, | |
| "loss": 3.2715, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 0.00093236402166235, | |
| "loss": 3.2625, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.3915330432954704, | |
| "eval_loss": 3.4678006172180176, | |
| "eval_runtime": 153.7826, | |
| "eval_samples_per_second": 376.629, | |
| "eval_steps_per_second": 5.885, | |
| "step": 55764 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 0.0009294207676006593, | |
| "loss": 3.2388, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 0.0009264775135389687, | |
| "loss": 3.1981, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "learning_rate": 0.000923534259477278, | |
| "loss": 3.1952, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 0.0009205939486696492, | |
| "loss": 3.1991, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 0.0009176506946079586, | |
| "loss": 3.2015, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 0.0009147074405462679, | |
| "loss": 3.1977, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 3.34, | |
| "learning_rate": 0.0009117641864845774, | |
| "loss": 3.1985, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 0.0009088238756769485, | |
| "loss": 3.195, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 0.0009058835648693195, | |
| "loss": 3.1988, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 0.000902940310807629, | |
| "loss": 3.2016, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 0.0008999970567459384, | |
| "loss": 3.1942, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 0.0008970538026842477, | |
| "loss": 3.1928, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 0.0008941134918766188, | |
| "loss": 3.1918, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 0.0008911702378149283, | |
| "loss": 3.1904, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 0.0008882269837532376, | |
| "loss": 3.1905, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 0.0008852866729456087, | |
| "loss": 3.191, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 0.0008823434188839181, | |
| "loss": 3.1849, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 0.0008794031080762892, | |
| "loss": 3.1891, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 3.98, | |
| "learning_rate": 0.0008764598540145986, | |
| "loss": 3.1838, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.3997996531181377, | |
| "eval_loss": 3.3962149620056152, | |
| "eval_runtime": 153.4429, | |
| "eval_samples_per_second": 377.463, | |
| "eval_steps_per_second": 5.898, | |
| "step": 74352 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 0.0008735195432069696, | |
| "loss": 3.1335, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 0.0008705762891452791, | |
| "loss": 3.1196, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 0.0008676359783376502, | |
| "loss": 3.1214, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "learning_rate": 0.0008646927242759594, | |
| "loss": 3.1243, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 0.0008617494702142689, | |
| "loss": 3.1284, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "learning_rate": 0.0008588062161525783, | |
| "loss": 3.1291, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 0.0008558629620908876, | |
| "loss": 3.1335, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 0.0008529226512832588, | |
| "loss": 3.1305, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 0.0008499793972215682, | |
| "loss": 3.1299, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 0.0008470390864139392, | |
| "loss": 3.1297, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 0.0008440958323522487, | |
| "loss": 3.1277, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "learning_rate": 0.0008411555215446197, | |
| "loss": 3.1316, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "learning_rate": 0.0008382122674829291, | |
| "loss": 3.126, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 0.0008352690134212385, | |
| "loss": 3.1336, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 0.0008323257593595479, | |
| "loss": 3.1283, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 4.84, | |
| "learning_rate": 0.0008293825052978573, | |
| "loss": 3.1309, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 0.0008264421944902284, | |
| "loss": 3.1292, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 0.0008234989404285378, | |
| "loss": 3.1277, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.4017150477760334, | |
| "eval_loss": 3.3849174976348877, | |
| "eval_runtime": 153.5044, | |
| "eval_samples_per_second": 377.312, | |
| "eval_steps_per_second": 5.896, | |
| "step": 92940 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0008205556863668472, | |
| "loss": 3.1209, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 0.0008176153755592183, | |
| "loss": 3.0617, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 0.0008146721214975277, | |
| "loss": 3.0649, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 0.0008117288674358371, | |
| "loss": 3.0697, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 0.0008087885566282081, | |
| "loss": 3.0744, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 5.27, | |
| "learning_rate": 0.0008058482458205792, | |
| "loss": 3.0762, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 5.33, | |
| "learning_rate": 0.0008029049917588887, | |
| "loss": 3.0775, | |
| "step": 99000 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 0.000799961737697198, | |
| "loss": 3.074, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "learning_rate": 0.0007970214268895691, | |
| "loss": 3.0827, | |
| "step": 101000 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "learning_rate": 0.0007940781728278786, | |
| "loss": 3.0798, | |
| "step": 102000 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "learning_rate": 0.0007911349187661879, | |
| "loss": 3.0802, | |
| "step": 103000 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 0.0007881916647044973, | |
| "loss": 3.0819, | |
| "step": 104000 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "learning_rate": 0.0007852484106428068, | |
| "loss": 3.0867, | |
| "step": 105000 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "learning_rate": 0.0007823051565811161, | |
| "loss": 3.081, | |
| "step": 106000 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 0.0007793648457734872, | |
| "loss": 3.0817, | |
| "step": 107000 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "learning_rate": 0.00077642747821992, | |
| "loss": 3.0864, | |
| "step": 108000 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 0.0007734842241582294, | |
| "loss": 3.0817, | |
| "step": 109000 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "learning_rate": 0.0007705409700965388, | |
| "loss": 3.0866, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "learning_rate": 0.0007675977160348481, | |
| "loss": 3.0813, | |
| "step": 111000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.40404998715837087, | |
| "eval_loss": 3.387449026107788, | |
| "eval_runtime": 153.7852, | |
| "eval_samples_per_second": 376.623, | |
| "eval_steps_per_second": 5.885, | |
| "step": 111528 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 0.0007646544619731576, | |
| "loss": 3.0506, | |
| "step": 112000 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 0.000761711207911467, | |
| "loss": 3.0138, | |
| "step": 113000 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "learning_rate": 0.000758770897103838, | |
| "loss": 3.0205, | |
| "step": 114000 | |
| }, | |
| { | |
| "epoch": 6.19, | |
| "learning_rate": 0.0007558276430421475, | |
| "loss": 3.0302, | |
| "step": 115000 | |
| }, | |
| { | |
| "epoch": 6.24, | |
| "learning_rate": 0.0007528873322345186, | |
| "loss": 3.0285, | |
| "step": 116000 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "learning_rate": 0.0007499440781728279, | |
| "loss": 3.0317, | |
| "step": 117000 | |
| }, | |
| { | |
| "epoch": 6.35, | |
| "learning_rate": 0.0007470008241111372, | |
| "loss": 3.0377, | |
| "step": 118000 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 0.0007440575700494467, | |
| "loss": 3.0375, | |
| "step": 119000 | |
| }, | |
| { | |
| "epoch": 6.46, | |
| "learning_rate": 0.0007411172592418177, | |
| "loss": 3.0414, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 6.51, | |
| "learning_rate": 0.0007381740051801271, | |
| "loss": 3.0435, | |
| "step": 121000 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 0.0007352307511184366, | |
| "loss": 3.0441, | |
| "step": 122000 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 0.0007322874970567459, | |
| "loss": 3.0481, | |
| "step": 123000 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 0.000729347186249117, | |
| "loss": 3.0422, | |
| "step": 124000 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "learning_rate": 0.0007264039321874263, | |
| "loss": 3.0457, | |
| "step": 125000 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "learning_rate": 0.0007234606781257358, | |
| "loss": 3.0457, | |
| "step": 126000 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 0.0007205203673181069, | |
| "loss": 3.0501, | |
| "step": 127000 | |
| }, | |
| { | |
| "epoch": 6.89, | |
| "learning_rate": 0.0007175771132564162, | |
| "loss": 3.052, | |
| "step": 128000 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "learning_rate": 0.0007146368024487874, | |
| "loss": 3.0474, | |
| "step": 129000 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "learning_rate": 0.0007116935483870968, | |
| "loss": 3.0519, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.40786908699745245, | |
| "eval_loss": 3.3393681049346924, | |
| "eval_runtime": 153.4305, | |
| "eval_samples_per_second": 377.493, | |
| "eval_steps_per_second": 5.898, | |
| "step": 130116 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 0.0007087532375794678, | |
| "loss": 2.9879, | |
| "step": 131000 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 0.000705812926771839, | |
| "loss": 2.984, | |
| "step": 132000 | |
| }, | |
| { | |
| "epoch": 7.16, | |
| "learning_rate": 0.0007028696727101484, | |
| "loss": 2.9883, | |
| "step": 133000 | |
| }, | |
| { | |
| "epoch": 7.21, | |
| "learning_rate": 0.0006999264186484577, | |
| "loss": 2.9895, | |
| "step": 134000 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "learning_rate": 0.0006969831645867672, | |
| "loss": 3.0, | |
| "step": 135000 | |
| }, | |
| { | |
| "epoch": 7.32, | |
| "learning_rate": 0.0006940399105250766, | |
| "loss": 3.0035, | |
| "step": 136000 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 0.0006910966564633859, | |
| "loss": 3.0073, | |
| "step": 137000 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "learning_rate": 0.000688156345655757, | |
| "loss": 3.0078, | |
| "step": 138000 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "learning_rate": 0.0006852160348481281, | |
| "loss": 3.0042, | |
| "step": 139000 | |
| }, | |
| { | |
| "epoch": 7.53, | |
| "learning_rate": 0.0006822757240404992, | |
| "loss": 3.0086, | |
| "step": 140000 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "learning_rate": 0.0006793324699788086, | |
| "loss": 3.0068, | |
| "step": 141000 | |
| }, | |
| { | |
| "epoch": 7.64, | |
| "learning_rate": 0.000676389215917118, | |
| "loss": 3.0091, | |
| "step": 142000 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "learning_rate": 0.0006734459618554274, | |
| "loss": 3.0145, | |
| "step": 143000 | |
| }, | |
| { | |
| "epoch": 7.75, | |
| "learning_rate": 0.0006705027077937368, | |
| "loss": 3.015, | |
| "step": 144000 | |
| }, | |
| { | |
| "epoch": 7.8, | |
| "learning_rate": 0.0006675623969861078, | |
| "loss": 3.0151, | |
| "step": 145000 | |
| }, | |
| { | |
| "epoch": 7.85, | |
| "learning_rate": 0.0006646191429244173, | |
| "loss": 3.0152, | |
| "step": 146000 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "learning_rate": 0.0006616788321167884, | |
| "loss": 3.0181, | |
| "step": 147000 | |
| }, | |
| { | |
| "epoch": 7.96, | |
| "learning_rate": 0.0006587355780550977, | |
| "loss": 3.0181, | |
| "step": 148000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.40853820876773905, | |
| "eval_loss": 3.344135284423828, | |
| "eval_runtime": 153.5235, | |
| "eval_samples_per_second": 377.265, | |
| "eval_steps_per_second": 5.895, | |
| "step": 148704 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 0.0006557952672474689, | |
| "loss": 2.9971, | |
| "step": 149000 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 0.0006528520131857782, | |
| "loss": 2.9484, | |
| "step": 150000 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "learning_rate": 0.0006499117023781493, | |
| "loss": 2.9536, | |
| "step": 151000 | |
| }, | |
| { | |
| "epoch": 8.18, | |
| "learning_rate": 0.0006469684483164588, | |
| "loss": 2.9635, | |
| "step": 152000 | |
| }, | |
| { | |
| "epoch": 8.23, | |
| "learning_rate": 0.0006440251942547681, | |
| "loss": 2.967, | |
| "step": 153000 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "learning_rate": 0.0006410819401930775, | |
| "loss": 2.9704, | |
| "step": 154000 | |
| }, | |
| { | |
| "epoch": 8.34, | |
| "learning_rate": 0.000638138686131387, | |
| "loss": 2.9718, | |
| "step": 155000 | |
| }, | |
| { | |
| "epoch": 8.39, | |
| "learning_rate": 0.0006351954320696963, | |
| "loss": 2.9766, | |
| "step": 156000 | |
| }, | |
| { | |
| "epoch": 8.45, | |
| "learning_rate": 0.0006322551212620673, | |
| "loss": 2.9788, | |
| "step": 157000 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "learning_rate": 0.0006293118672003767, | |
| "loss": 2.9812, | |
| "step": 158000 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "learning_rate": 0.0006263715563927478, | |
| "loss": 2.9796, | |
| "step": 159000 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "learning_rate": 0.0006234283023310572, | |
| "loss": 2.9815, | |
| "step": 160000 | |
| }, | |
| { | |
| "epoch": 8.66, | |
| "learning_rate": 0.0006204879915234283, | |
| "loss": 2.9841, | |
| "step": 161000 | |
| }, | |
| { | |
| "epoch": 8.72, | |
| "learning_rate": 0.0006175447374617377, | |
| "loss": 2.9853, | |
| "step": 162000 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "learning_rate": 0.0006146044266541088, | |
| "loss": 2.9876, | |
| "step": 163000 | |
| }, | |
| { | |
| "epoch": 8.82, | |
| "learning_rate": 0.0006116611725924181, | |
| "loss": 2.9846, | |
| "step": 164000 | |
| }, | |
| { | |
| "epoch": 8.88, | |
| "learning_rate": 0.0006087179185307275, | |
| "loss": 2.9885, | |
| "step": 165000 | |
| }, | |
| { | |
| "epoch": 8.93, | |
| "learning_rate": 0.000605774664469037, | |
| "loss": 2.985, | |
| "step": 166000 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "learning_rate": 0.000602834353661408, | |
| "loss": 2.9888, | |
| "step": 167000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.40875567334308216, | |
| "eval_loss": 3.3545050621032715, | |
| "eval_runtime": 153.9524, | |
| "eval_samples_per_second": 376.214, | |
| "eval_steps_per_second": 5.878, | |
| "step": 167292 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 0.0005998940428537791, | |
| "loss": 2.9412, | |
| "step": 168000 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 0.0005969507887920886, | |
| "loss": 2.9225, | |
| "step": 169000 | |
| }, | |
| { | |
| "epoch": 9.15, | |
| "learning_rate": 0.0005940104779844596, | |
| "loss": 2.9315, | |
| "step": 170000 | |
| }, | |
| { | |
| "epoch": 9.2, | |
| "learning_rate": 0.000591067223922769, | |
| "loss": 2.9394, | |
| "step": 171000 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "learning_rate": 0.0005881239698610785, | |
| "loss": 2.9384, | |
| "step": 172000 | |
| }, | |
| { | |
| "epoch": 9.31, | |
| "learning_rate": 0.0005851807157993878, | |
| "loss": 2.9457, | |
| "step": 173000 | |
| }, | |
| { | |
| "epoch": 9.36, | |
| "learning_rate": 0.0005822404049917589, | |
| "loss": 2.9486, | |
| "step": 174000 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "learning_rate": 0.0005792971509300682, | |
| "loss": 2.9479, | |
| "step": 175000 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "learning_rate": 0.0005763538968683777, | |
| "loss": 2.9523, | |
| "step": 176000 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 0.0005734106428066871, | |
| "loss": 2.9536, | |
| "step": 177000 | |
| }, | |
| { | |
| "epoch": 9.58, | |
| "learning_rate": 0.0005704703319990581, | |
| "loss": 2.9541, | |
| "step": 178000 | |
| }, | |
| { | |
| "epoch": 9.63, | |
| "learning_rate": 0.0005675300211914293, | |
| "loss": 2.9613, | |
| "step": 179000 | |
| }, | |
| { | |
| "epoch": 9.68, | |
| "learning_rate": 0.0005645867671297387, | |
| "loss": 2.9567, | |
| "step": 180000 | |
| }, | |
| { | |
| "epoch": 9.74, | |
| "learning_rate": 0.0005616464563221097, | |
| "loss": 2.9606, | |
| "step": 181000 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "learning_rate": 0.0005587032022604192, | |
| "loss": 2.9578, | |
| "step": 182000 | |
| }, | |
| { | |
| "epoch": 9.85, | |
| "learning_rate": 0.000555765834706852, | |
| "loss": 2.9623, | |
| "step": 183000 | |
| }, | |
| { | |
| "epoch": 9.9, | |
| "learning_rate": 0.0005528225806451613, | |
| "loss": 2.959, | |
| "step": 184000 | |
| }, | |
| { | |
| "epoch": 9.95, | |
| "learning_rate": 0.0005498793265834707, | |
| "loss": 2.9602, | |
| "step": 185000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.4088348124440257, | |
| "eval_loss": 3.3501293659210205, | |
| "eval_runtime": 153.6667, | |
| "eval_samples_per_second": 376.913, | |
| "eval_steps_per_second": 5.889, | |
| "step": 185880 | |
| }, | |
| { | |
| "epoch": 10.01, | |
| "learning_rate": 0.0005469360725217802, | |
| "loss": 2.9596, | |
| "step": 186000 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "learning_rate": 0.0005439928184600895, | |
| "loss": 2.9011, | |
| "step": 187000 | |
| }, | |
| { | |
| "epoch": 10.11, | |
| "learning_rate": 0.0005410495643983989, | |
| "loss": 2.9033, | |
| "step": 188000 | |
| }, | |
| { | |
| "epoch": 10.17, | |
| "learning_rate": 0.00053810925359077, | |
| "loss": 2.9104, | |
| "step": 189000 | |
| }, | |
| { | |
| "epoch": 10.22, | |
| "learning_rate": 0.0005351689427831411, | |
| "loss": 2.9175, | |
| "step": 190000 | |
| }, | |
| { | |
| "epoch": 10.28, | |
| "learning_rate": 0.0005322256887214505, | |
| "loss": 2.9182, | |
| "step": 191000 | |
| }, | |
| { | |
| "epoch": 10.33, | |
| "learning_rate": 0.0005292824346597599, | |
| "loss": 2.9185, | |
| "step": 192000 | |
| }, | |
| { | |
| "epoch": 10.38, | |
| "learning_rate": 0.0005263421238521309, | |
| "loss": 2.921, | |
| "step": 193000 | |
| }, | |
| { | |
| "epoch": 10.44, | |
| "learning_rate": 0.0005233988697904403, | |
| "loss": 2.9291, | |
| "step": 194000 | |
| }, | |
| { | |
| "epoch": 10.49, | |
| "learning_rate": 0.0005204585589828113, | |
| "loss": 2.9294, | |
| "step": 195000 | |
| }, | |
| { | |
| "epoch": 10.54, | |
| "learning_rate": 0.0005175153049211208, | |
| "loss": 2.9324, | |
| "step": 196000 | |
| }, | |
| { | |
| "epoch": 10.6, | |
| "learning_rate": 0.0005145749941134919, | |
| "loss": 2.9271, | |
| "step": 197000 | |
| }, | |
| { | |
| "epoch": 10.65, | |
| "learning_rate": 0.0005116317400518012, | |
| "loss": 2.9325, | |
| "step": 198000 | |
| }, | |
| { | |
| "epoch": 10.71, | |
| "learning_rate": 0.0005086884859901107, | |
| "loss": 2.9341, | |
| "step": 199000 | |
| }, | |
| { | |
| "epoch": 10.76, | |
| "learning_rate": 0.0005057481751824817, | |
| "loss": 2.9395, | |
| "step": 200000 | |
| }, | |
| { | |
| "epoch": 10.81, | |
| "learning_rate": 0.0005028049211207911, | |
| "loss": 2.9351, | |
| "step": 201000 | |
| }, | |
| { | |
| "epoch": 10.87, | |
| "learning_rate": 0.0004998616670591005, | |
| "loss": 2.9344, | |
| "step": 202000 | |
| }, | |
| { | |
| "epoch": 10.92, | |
| "learning_rate": 0.0004969213562514716, | |
| "loss": 2.9385, | |
| "step": 203000 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "learning_rate": 0.000493978102189781, | |
| "loss": 2.942, | |
| "step": 204000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.40947598695964976, | |
| "eval_loss": 3.35093092918396, | |
| "eval_runtime": 153.6805, | |
| "eval_samples_per_second": 376.879, | |
| "eval_steps_per_second": 5.889, | |
| "step": 204468 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 0.0004910377913821521, | |
| "loss": 2.9079, | |
| "step": 205000 | |
| }, | |
| { | |
| "epoch": 11.08, | |
| "learning_rate": 0.0004880974805745232, | |
| "loss": 2.881, | |
| "step": 206000 | |
| }, | |
| { | |
| "epoch": 11.14, | |
| "learning_rate": 0.0004851542265128326, | |
| "loss": 2.8806, | |
| "step": 207000 | |
| }, | |
| { | |
| "epoch": 11.19, | |
| "learning_rate": 0.00048221391570520367, | |
| "loss": 2.8891, | |
| "step": 208000 | |
| }, | |
| { | |
| "epoch": 11.24, | |
| "learning_rate": 0.00047927066164351305, | |
| "loss": 2.8992, | |
| "step": 209000 | |
| }, | |
| { | |
| "epoch": 11.3, | |
| "learning_rate": 0.0004763303508358842, | |
| "loss": 2.8893, | |
| "step": 210000 | |
| }, | |
| { | |
| "epoch": 11.35, | |
| "learning_rate": 0.00047338709677419356, | |
| "loss": 2.9037, | |
| "step": 211000 | |
| }, | |
| { | |
| "epoch": 11.41, | |
| "learning_rate": 0.00047044384271250294, | |
| "loss": 2.9027, | |
| "step": 212000 | |
| }, | |
| { | |
| "epoch": 11.46, | |
| "learning_rate": 0.00046750058865081237, | |
| "loss": 2.9063, | |
| "step": 213000 | |
| }, | |
| { | |
| "epoch": 11.51, | |
| "learning_rate": 0.00046455733458912175, | |
| "loss": 2.905, | |
| "step": 214000 | |
| }, | |
| { | |
| "epoch": 11.57, | |
| "learning_rate": 0.00046161408052743113, | |
| "loss": 2.9073, | |
| "step": 215000 | |
| }, | |
| { | |
| "epoch": 11.62, | |
| "learning_rate": 0.0004586737697198022, | |
| "loss": 2.9115, | |
| "step": 216000 | |
| }, | |
| { | |
| "epoch": 11.67, | |
| "learning_rate": 0.00045573345891217334, | |
| "loss": 2.9121, | |
| "step": 217000 | |
| }, | |
| { | |
| "epoch": 11.73, | |
| "learning_rate": 0.0004527902048504827, | |
| "loss": 2.9119, | |
| "step": 218000 | |
| }, | |
| { | |
| "epoch": 11.78, | |
| "learning_rate": 0.0004498498940428538, | |
| "loss": 2.9123, | |
| "step": 219000 | |
| }, | |
| { | |
| "epoch": 11.84, | |
| "learning_rate": 0.00044690663998116323, | |
| "loss": 2.9137, | |
| "step": 220000 | |
| }, | |
| { | |
| "epoch": 11.89, | |
| "learning_rate": 0.00044396632917353425, | |
| "loss": 2.9182, | |
| "step": 221000 | |
| }, | |
| { | |
| "epoch": 11.94, | |
| "learning_rate": 0.00044102307511184363, | |
| "loss": 2.9199, | |
| "step": 222000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 0.000438079821050153, | |
| "loss": 2.9174, | |
| "step": 223000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.4092696072329107, | |
| "eval_loss": 3.370931625366211, | |
| "eval_runtime": 153.6062, | |
| "eval_samples_per_second": 377.061, | |
| "eval_steps_per_second": 5.892, | |
| "step": 223056 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 0.00043513951024252414, | |
| "loss": 2.8618, | |
| "step": 224000 | |
| }, | |
| { | |
| "epoch": 12.1, | |
| "learning_rate": 0.0004321962561808335, | |
| "loss": 2.8657, | |
| "step": 225000 | |
| }, | |
| { | |
| "epoch": 12.16, | |
| "learning_rate": 0.0004292559453732046, | |
| "loss": 2.8649, | |
| "step": 226000 | |
| }, | |
| { | |
| "epoch": 12.21, | |
| "learning_rate": 0.00042631269131151404, | |
| "loss": 2.8739, | |
| "step": 227000 | |
| }, | |
| { | |
| "epoch": 12.27, | |
| "learning_rate": 0.0004233694372498234, | |
| "loss": 2.8766, | |
| "step": 228000 | |
| }, | |
| { | |
| "epoch": 12.32, | |
| "learning_rate": 0.0004204261831881328, | |
| "loss": 2.8788, | |
| "step": 229000 | |
| }, | |
| { | |
| "epoch": 12.37, | |
| "learning_rate": 0.00041748881563456557, | |
| "loss": 2.8791, | |
| "step": 230000 | |
| }, | |
| { | |
| "epoch": 12.43, | |
| "learning_rate": 0.000414545561572875, | |
| "loss": 2.8812, | |
| "step": 231000 | |
| }, | |
| { | |
| "epoch": 12.48, | |
| "learning_rate": 0.0004116023075111844, | |
| "loss": 2.8795, | |
| "step": 232000 | |
| }, | |
| { | |
| "epoch": 12.53, | |
| "learning_rate": 0.00040866493995761716, | |
| "loss": 2.8892, | |
| "step": 233000 | |
| }, | |
| { | |
| "epoch": 12.59, | |
| "learning_rate": 0.00040572168589592654, | |
| "loss": 2.8891, | |
| "step": 234000 | |
| }, | |
| { | |
| "epoch": 12.64, | |
| "learning_rate": 0.00040277843183423597, | |
| "loss": 2.8872, | |
| "step": 235000 | |
| }, | |
| { | |
| "epoch": 12.7, | |
| "learning_rate": 0.00039983517777254535, | |
| "loss": 2.8884, | |
| "step": 236000 | |
| }, | |
| { | |
| "epoch": 12.75, | |
| "learning_rate": 0.00039689192371085473, | |
| "loss": 2.8897, | |
| "step": 237000 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "learning_rate": 0.00039395161290322586, | |
| "loss": 2.8915, | |
| "step": 238000 | |
| }, | |
| { | |
| "epoch": 12.86, | |
| "learning_rate": 0.0003910083588415352, | |
| "loss": 2.8951, | |
| "step": 239000 | |
| }, | |
| { | |
| "epoch": 12.91, | |
| "learning_rate": 0.00038806804803390626, | |
| "loss": 2.8917, | |
| "step": 240000 | |
| }, | |
| { | |
| "epoch": 12.97, | |
| "learning_rate": 0.00038512479397221565, | |
| "loss": 2.8989, | |
| "step": 241000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.4106622001220946, | |
| "eval_loss": 3.3607850074768066, | |
| "eval_runtime": 153.7023, | |
| "eval_samples_per_second": 376.826, | |
| "eval_steps_per_second": 5.888, | |
| "step": 241644 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "learning_rate": 0.0003821844831645868, | |
| "loss": 2.8762, | |
| "step": 242000 | |
| }, | |
| { | |
| "epoch": 13.07, | |
| "learning_rate": 0.00037924122910289616, | |
| "loss": 2.8394, | |
| "step": 243000 | |
| }, | |
| { | |
| "epoch": 13.13, | |
| "learning_rate": 0.00037629797504120554, | |
| "loss": 2.8468, | |
| "step": 244000 | |
| }, | |
| { | |
| "epoch": 13.18, | |
| "learning_rate": 0.00037335766423357667, | |
| "loss": 2.8509, | |
| "step": 245000 | |
| }, | |
| { | |
| "epoch": 13.23, | |
| "learning_rate": 0.00037041441017188605, | |
| "loss": 2.8549, | |
| "step": 246000 | |
| }, | |
| { | |
| "epoch": 13.29, | |
| "learning_rate": 0.00036747115611019543, | |
| "loss": 2.8499, | |
| "step": 247000 | |
| }, | |
| { | |
| "epoch": 13.34, | |
| "learning_rate": 0.0003645337885566282, | |
| "loss": 2.8571, | |
| "step": 248000 | |
| }, | |
| { | |
| "epoch": 13.4, | |
| "learning_rate": 0.00036159053449493763, | |
| "loss": 2.8601, | |
| "step": 249000 | |
| }, | |
| { | |
| "epoch": 13.45, | |
| "learning_rate": 0.000358647280433247, | |
| "loss": 2.8638, | |
| "step": 250000 | |
| }, | |
| { | |
| "epoch": 13.5, | |
| "learning_rate": 0.0003557040263715564, | |
| "loss": 2.8653, | |
| "step": 251000 | |
| }, | |
| { | |
| "epoch": 13.56, | |
| "learning_rate": 0.0003527637155639275, | |
| "loss": 2.8664, | |
| "step": 252000 | |
| }, | |
| { | |
| "epoch": 13.61, | |
| "learning_rate": 0.0003498204615022369, | |
| "loss": 2.8661, | |
| "step": 253000 | |
| }, | |
| { | |
| "epoch": 13.66, | |
| "learning_rate": 0.000346880150694608, | |
| "loss": 2.8729, | |
| "step": 254000 | |
| }, | |
| { | |
| "epoch": 13.72, | |
| "learning_rate": 0.00034393689663291736, | |
| "loss": 2.8698, | |
| "step": 255000 | |
| }, | |
| { | |
| "epoch": 13.77, | |
| "learning_rate": 0.0003409936425712268, | |
| "loss": 2.875, | |
| "step": 256000 | |
| }, | |
| { | |
| "epoch": 13.83, | |
| "learning_rate": 0.0003380503885095362, | |
| "loss": 2.873, | |
| "step": 257000 | |
| }, | |
| { | |
| "epoch": 13.88, | |
| "learning_rate": 0.0003351071344478455, | |
| "loss": 2.8736, | |
| "step": 258000 | |
| }, | |
| { | |
| "epoch": 13.93, | |
| "learning_rate": 0.00033216388038615494, | |
| "loss": 2.8746, | |
| "step": 259000 | |
| }, | |
| { | |
| "epoch": 13.99, | |
| "learning_rate": 0.000329223569578526, | |
| "loss": 2.8757, | |
| "step": 260000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.41005945308163155, | |
| "eval_loss": 3.3651070594787598, | |
| "eval_runtime": 153.628, | |
| "eval_samples_per_second": 377.008, | |
| "eval_steps_per_second": 5.891, | |
| "step": 260232 | |
| }, | |
| { | |
| "epoch": 14.04, | |
| "learning_rate": 0.0003262803155168354, | |
| "loss": 2.8343, | |
| "step": 261000 | |
| }, | |
| { | |
| "epoch": 14.1, | |
| "learning_rate": 0.00032334000470920647, | |
| "loss": 2.8301, | |
| "step": 262000 | |
| }, | |
| { | |
| "epoch": 14.15, | |
| "learning_rate": 0.0003203967506475159, | |
| "loss": 2.8292, | |
| "step": 263000 | |
| }, | |
| { | |
| "epoch": 14.2, | |
| "learning_rate": 0.0003174534965858253, | |
| "loss": 2.8317, | |
| "step": 264000 | |
| }, | |
| { | |
| "epoch": 14.26, | |
| "learning_rate": 0.00031451318577819636, | |
| "loss": 2.8366, | |
| "step": 265000 | |
| }, | |
| { | |
| "epoch": 14.31, | |
| "learning_rate": 0.00031157287497056744, | |
| "loss": 2.8386, | |
| "step": 266000 | |
| }, | |
| { | |
| "epoch": 14.36, | |
| "learning_rate": 0.0003086296209088769, | |
| "loss": 2.8376, | |
| "step": 267000 | |
| }, | |
| { | |
| "epoch": 14.42, | |
| "learning_rate": 0.00030568636684718625, | |
| "loss": 2.8456, | |
| "step": 268000 | |
| }, | |
| { | |
| "epoch": 14.47, | |
| "learning_rate": 0.00030274605603955733, | |
| "loss": 2.8454, | |
| "step": 269000 | |
| }, | |
| { | |
| "epoch": 14.53, | |
| "learning_rate": 0.00029980280197786677, | |
| "loss": 2.8477, | |
| "step": 270000 | |
| }, | |
| { | |
| "epoch": 14.58, | |
| "learning_rate": 0.00029685954791617615, | |
| "loss": 2.8442, | |
| "step": 271000 | |
| }, | |
| { | |
| "epoch": 14.63, | |
| "learning_rate": 0.0002939162938544855, | |
| "loss": 2.8486, | |
| "step": 272000 | |
| }, | |
| { | |
| "epoch": 14.69, | |
| "learning_rate": 0.00029097303979279496, | |
| "loss": 2.8506, | |
| "step": 273000 | |
| }, | |
| { | |
| "epoch": 14.74, | |
| "learning_rate": 0.00028802978573110434, | |
| "loss": 2.8504, | |
| "step": 274000 | |
| }, | |
| { | |
| "epoch": 14.79, | |
| "learning_rate": 0.0002850865316694137, | |
| "loss": 2.8578, | |
| "step": 275000 | |
| }, | |
| { | |
| "epoch": 14.85, | |
| "learning_rate": 0.0002821462208617848, | |
| "loss": 2.8522, | |
| "step": 276000 | |
| }, | |
| { | |
| "epoch": 14.9, | |
| "learning_rate": 0.0002792059100541559, | |
| "loss": 2.8574, | |
| "step": 277000 | |
| }, | |
| { | |
| "epoch": 14.96, | |
| "learning_rate": 0.00027626265599246526, | |
| "loss": 2.8506, | |
| "step": 278000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.41093038627741424, | |
| "eval_loss": 3.363818407058716, | |
| "eval_runtime": 153.8072, | |
| "eval_samples_per_second": 376.569, | |
| "eval_steps_per_second": 5.884, | |
| "step": 278820 | |
| }, | |
| { | |
| "epoch": 15.01, | |
| "learning_rate": 0.00027332234518483633, | |
| "loss": 2.8477, | |
| "step": 279000 | |
| }, | |
| { | |
| "epoch": 15.06, | |
| "learning_rate": 0.0002703790911231457, | |
| "loss": 2.8029, | |
| "step": 280000 | |
| }, | |
| { | |
| "epoch": 15.12, | |
| "learning_rate": 0.00026743878031551684, | |
| "loss": 2.8113, | |
| "step": 281000 | |
| }, | |
| { | |
| "epoch": 15.17, | |
| "learning_rate": 0.0002644955262538262, | |
| "loss": 2.8192, | |
| "step": 282000 | |
| }, | |
| { | |
| "epoch": 15.22, | |
| "learning_rate": 0.0002615522721921356, | |
| "loss": 2.8172, | |
| "step": 283000 | |
| }, | |
| { | |
| "epoch": 15.28, | |
| "learning_rate": 0.00025860901813044504, | |
| "loss": 2.8204, | |
| "step": 284000 | |
| }, | |
| { | |
| "epoch": 15.33, | |
| "learning_rate": 0.0002556687073228161, | |
| "loss": 2.8208, | |
| "step": 285000 | |
| }, | |
| { | |
| "epoch": 15.39, | |
| "learning_rate": 0.0002527254532611255, | |
| "loss": 2.8246, | |
| "step": 286000 | |
| }, | |
| { | |
| "epoch": 15.44, | |
| "learning_rate": 0.0002497821991994349, | |
| "loss": 2.8226, | |
| "step": 287000 | |
| }, | |
| { | |
| "epoch": 15.49, | |
| "learning_rate": 0.000246841888391806, | |
| "loss": 2.8279, | |
| "step": 288000 | |
| }, | |
| { | |
| "epoch": 15.55, | |
| "learning_rate": 0.00024389863433011539, | |
| "loss": 2.8303, | |
| "step": 289000 | |
| }, | |
| { | |
| "epoch": 15.6, | |
| "learning_rate": 0.0002409553802684248, | |
| "loss": 2.828, | |
| "step": 290000 | |
| }, | |
| { | |
| "epoch": 15.66, | |
| "learning_rate": 0.00023801506946079587, | |
| "loss": 2.829, | |
| "step": 291000 | |
| }, | |
| { | |
| "epoch": 15.71, | |
| "learning_rate": 0.00023507181539910525, | |
| "loss": 2.8336, | |
| "step": 292000 | |
| }, | |
| { | |
| "epoch": 15.76, | |
| "learning_rate": 0.00023212856133741463, | |
| "loss": 2.8294, | |
| "step": 293000 | |
| }, | |
| { | |
| "epoch": 15.82, | |
| "learning_rate": 0.00022918825052978573, | |
| "loss": 2.8337, | |
| "step": 294000 | |
| }, | |
| { | |
| "epoch": 15.87, | |
| "learning_rate": 0.00022624499646809512, | |
| "loss": 2.837, | |
| "step": 295000 | |
| }, | |
| { | |
| "epoch": 15.92, | |
| "learning_rate": 0.00022330174240640452, | |
| "loss": 2.8345, | |
| "step": 296000 | |
| }, | |
| { | |
| "epoch": 15.98, | |
| "learning_rate": 0.00022035848834471393, | |
| "loss": 2.8373, | |
| "step": 297000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.41065266043420495, | |
| "eval_loss": 3.372394561767578, | |
| "eval_runtime": 153.6132, | |
| "eval_samples_per_second": 377.044, | |
| "eval_steps_per_second": 5.891, | |
| "step": 297408 | |
| }, | |
| { | |
| "epoch": 16.03, | |
| "learning_rate": 0.000217418177537085, | |
| "loss": 2.8089, | |
| "step": 298000 | |
| }, | |
| { | |
| "epoch": 16.09, | |
| "learning_rate": 0.0002144778667294561, | |
| "loss": 2.798, | |
| "step": 299000 | |
| }, | |
| { | |
| "epoch": 16.14, | |
| "learning_rate": 0.0002115346126677655, | |
| "loss": 2.793, | |
| "step": 300000 | |
| }, | |
| { | |
| "epoch": 16.19, | |
| "learning_rate": 0.0002085913586060749, | |
| "loss": 2.7981, | |
| "step": 301000 | |
| }, | |
| { | |
| "epoch": 16.25, | |
| "learning_rate": 0.00020565104779844595, | |
| "loss": 2.8002, | |
| "step": 302000 | |
| }, | |
| { | |
| "epoch": 16.3, | |
| "learning_rate": 0.00020270779373675535, | |
| "loss": 2.8011, | |
| "step": 303000 | |
| }, | |
| { | |
| "epoch": 16.35, | |
| "learning_rate": 0.00019976748292912643, | |
| "loss": 2.8051, | |
| "step": 304000 | |
| }, | |
| { | |
| "epoch": 16.41, | |
| "learning_rate": 0.00019682422886743584, | |
| "loss": 2.8105, | |
| "step": 305000 | |
| }, | |
| { | |
| "epoch": 16.46, | |
| "learning_rate": 0.00019388097480574525, | |
| "loss": 2.8114, | |
| "step": 306000 | |
| }, | |
| { | |
| "epoch": 16.52, | |
| "learning_rate": 0.00019093772074405463, | |
| "loss": 2.8076, | |
| "step": 307000 | |
| }, | |
| { | |
| "epoch": 16.57, | |
| "learning_rate": 0.00018799740993642573, | |
| "loss": 2.8102, | |
| "step": 308000 | |
| }, | |
| { | |
| "epoch": 16.62, | |
| "learning_rate": 0.0001850541558747351, | |
| "loss": 2.8088, | |
| "step": 309000 | |
| }, | |
| { | |
| "epoch": 16.68, | |
| "learning_rate": 0.00018211090181304452, | |
| "loss": 2.8144, | |
| "step": 310000 | |
| }, | |
| { | |
| "epoch": 16.73, | |
| "learning_rate": 0.00017917059100541557, | |
| "loss": 2.8158, | |
| "step": 311000 | |
| }, | |
| { | |
| "epoch": 16.79, | |
| "learning_rate": 0.00017622733694372497, | |
| "loss": 2.8143, | |
| "step": 312000 | |
| }, | |
| { | |
| "epoch": 16.84, | |
| "learning_rate": 0.00017328408288203438, | |
| "loss": 2.8121, | |
| "step": 313000 | |
| }, | |
| { | |
| "epoch": 16.89, | |
| "learning_rate": 0.00017034082882034376, | |
| "loss": 2.8186, | |
| "step": 314000 | |
| }, | |
| { | |
| "epoch": 16.95, | |
| "learning_rate": 0.00016740051801271487, | |
| "loss": 2.8195, | |
| "step": 315000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.4107544394986612, | |
| "eval_loss": 3.3818860054016113, | |
| "eval_runtime": 153.5625, | |
| "eval_samples_per_second": 377.169, | |
| "eval_steps_per_second": 5.893, | |
| "step": 315996 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 0.00016446020720508594, | |
| "loss": 2.8161, | |
| "step": 316000 | |
| }, | |
| { | |
| "epoch": 17.05, | |
| "learning_rate": 0.00016151695314339535, | |
| "loss": 2.775, | |
| "step": 317000 | |
| }, | |
| { | |
| "epoch": 17.11, | |
| "learning_rate": 0.00015857664233576643, | |
| "loss": 2.7802, | |
| "step": 318000 | |
| }, | |
| { | |
| "epoch": 17.16, | |
| "learning_rate": 0.00015563338827407583, | |
| "loss": 2.7816, | |
| "step": 319000 | |
| }, | |
| { | |
| "epoch": 17.22, | |
| "learning_rate": 0.00015269013421238524, | |
| "loss": 2.7826, | |
| "step": 320000 | |
| }, | |
| { | |
| "epoch": 17.27, | |
| "learning_rate": 0.0001497468801506946, | |
| "loss": 2.7866, | |
| "step": 321000 | |
| }, | |
| { | |
| "epoch": 17.32, | |
| "learning_rate": 0.000146803626089004, | |
| "loss": 2.7882, | |
| "step": 322000 | |
| }, | |
| { | |
| "epoch": 17.38, | |
| "learning_rate": 0.00014386331528137508, | |
| "loss": 2.7918, | |
| "step": 323000 | |
| }, | |
| { | |
| "epoch": 17.43, | |
| "learning_rate": 0.00014092006121968449, | |
| "loss": 2.7885, | |
| "step": 324000 | |
| }, | |
| { | |
| "epoch": 17.48, | |
| "learning_rate": 0.00013797680715799387, | |
| "loss": 2.7938, | |
| "step": 325000 | |
| }, | |
| { | |
| "epoch": 17.54, | |
| "learning_rate": 0.00013503355309630327, | |
| "loss": 2.7926, | |
| "step": 326000 | |
| }, | |
| { | |
| "epoch": 17.59, | |
| "learning_rate": 0.00013209324228867438, | |
| "loss": 2.7947, | |
| "step": 327000 | |
| }, | |
| { | |
| "epoch": 17.65, | |
| "learning_rate": 0.00012915293148104545, | |
| "loss": 2.7974, | |
| "step": 328000 | |
| }, | |
| { | |
| "epoch": 17.7, | |
| "learning_rate": 0.00012620967741935486, | |
| "loss": 2.7926, | |
| "step": 329000 | |
| }, | |
| { | |
| "epoch": 17.75, | |
| "learning_rate": 0.00012326642335766424, | |
| "loss": 2.7969, | |
| "step": 330000 | |
| }, | |
| { | |
| "epoch": 17.81, | |
| "learning_rate": 0.00012032611255003532, | |
| "loss": 2.797, | |
| "step": 331000 | |
| }, | |
| { | |
| "epoch": 17.86, | |
| "learning_rate": 0.00011738580174240641, | |
| "loss": 2.8026, | |
| "step": 332000 | |
| }, | |
| { | |
| "epoch": 17.91, | |
| "learning_rate": 0.0001144425476807158, | |
| "loss": 2.7987, | |
| "step": 333000 | |
| }, | |
| { | |
| "epoch": 17.97, | |
| "learning_rate": 0.0001114992936190252, | |
| "loss": 2.7983, | |
| "step": 334000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.41104412354218284, | |
| "eval_loss": 3.3819210529327393, | |
| "eval_runtime": 153.6749, | |
| "eval_samples_per_second": 376.893, | |
| "eval_steps_per_second": 5.889, | |
| "step": 334584 | |
| }, | |
| { | |
| "epoch": 18.02, | |
| "learning_rate": 0.00010855603955733459, | |
| "loss": 2.7861, | |
| "step": 335000 | |
| }, | |
| { | |
| "epoch": 18.08, | |
| "learning_rate": 0.00010561572874970568, | |
| "loss": 2.7654, | |
| "step": 336000 | |
| }, | |
| { | |
| "epoch": 18.13, | |
| "learning_rate": 0.00010267247468801507, | |
| "loss": 2.7679, | |
| "step": 337000 | |
| }, | |
| { | |
| "epoch": 18.18, | |
| "learning_rate": 9.972922062632445e-05, | |
| "loss": 2.772, | |
| "step": 338000 | |
| }, | |
| { | |
| "epoch": 18.24, | |
| "learning_rate": 9.678596656463386e-05, | |
| "loss": 2.7715, | |
| "step": 339000 | |
| }, | |
| { | |
| "epoch": 18.29, | |
| "learning_rate": 9.384565575700495e-05, | |
| "loss": 2.7719, | |
| "step": 340000 | |
| }, | |
| { | |
| "epoch": 18.35, | |
| "learning_rate": 9.090240169531435e-05, | |
| "loss": 2.7753, | |
| "step": 341000 | |
| }, | |
| { | |
| "epoch": 18.4, | |
| "learning_rate": 8.795914763362374e-05, | |
| "loss": 2.7716, | |
| "step": 342000 | |
| }, | |
| { | |
| "epoch": 18.45, | |
| "learning_rate": 8.501883682599482e-05, | |
| "loss": 2.7763, | |
| "step": 343000 | |
| }, | |
| { | |
| "epoch": 18.51, | |
| "learning_rate": 8.20785260183659e-05, | |
| "loss": 2.7768, | |
| "step": 344000 | |
| }, | |
| { | |
| "epoch": 18.56, | |
| "learning_rate": 7.91352719566753e-05, | |
| "loss": 2.7772, | |
| "step": 345000 | |
| }, | |
| { | |
| "epoch": 18.61, | |
| "learning_rate": 7.61920178949847e-05, | |
| "loss": 2.7777, | |
| "step": 346000 | |
| }, | |
| { | |
| "epoch": 18.67, | |
| "learning_rate": 7.32487638332941e-05, | |
| "loss": 2.7804, | |
| "step": 347000 | |
| }, | |
| { | |
| "epoch": 18.72, | |
| "learning_rate": 7.03055097716035e-05, | |
| "loss": 2.7769, | |
| "step": 348000 | |
| }, | |
| { | |
| "epoch": 18.78, | |
| "learning_rate": 6.736519896397457e-05, | |
| "loss": 2.777, | |
| "step": 349000 | |
| }, | |
| { | |
| "epoch": 18.83, | |
| "learning_rate": 6.442194490228397e-05, | |
| "loss": 2.7815, | |
| "step": 350000 | |
| }, | |
| { | |
| "epoch": 18.88, | |
| "learning_rate": 6.147869084059336e-05, | |
| "loss": 2.7791, | |
| "step": 351000 | |
| }, | |
| { | |
| "epoch": 18.94, | |
| "learning_rate": 5.853543677890276e-05, | |
| "loss": 2.7793, | |
| "step": 352000 | |
| }, | |
| { | |
| "epoch": 18.99, | |
| "learning_rate": 5.559512597127384e-05, | |
| "loss": 2.7786, | |
| "step": 353000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.4102747006872539, | |
| "eval_loss": 3.3970141410827637, | |
| "eval_runtime": 153.4893, | |
| "eval_samples_per_second": 377.349, | |
| "eval_steps_per_second": 5.896, | |
| "step": 353172 | |
| }, | |
| { | |
| "epoch": 19.04, | |
| "learning_rate": 5.2654815163644926e-05, | |
| "loss": 2.7643, | |
| "step": 354000 | |
| }, | |
| { | |
| "epoch": 19.1, | |
| "learning_rate": 4.971450435601601e-05, | |
| "loss": 2.7618, | |
| "step": 355000 | |
| }, | |
| { | |
| "epoch": 19.15, | |
| "learning_rate": 4.67712502943254e-05, | |
| "loss": 2.7583, | |
| "step": 356000 | |
| }, | |
| { | |
| "epoch": 19.21, | |
| "learning_rate": 4.3827996232634804e-05, | |
| "loss": 2.7595, | |
| "step": 357000 | |
| }, | |
| { | |
| "epoch": 19.26, | |
| "learning_rate": 4.088768542500589e-05, | |
| "loss": 2.7565, | |
| "step": 358000 | |
| }, | |
| { | |
| "epoch": 19.31, | |
| "learning_rate": 3.794443136331528e-05, | |
| "loss": 2.7626, | |
| "step": 359000 | |
| }, | |
| { | |
| "epoch": 19.37, | |
| "learning_rate": 3.500117730162468e-05, | |
| "loss": 2.7608, | |
| "step": 360000 | |
| }, | |
| { | |
| "epoch": 19.42, | |
| "learning_rate": 3.205792323993407e-05, | |
| "loss": 2.7597, | |
| "step": 361000 | |
| }, | |
| { | |
| "epoch": 19.47, | |
| "learning_rate": 2.9114669178243463e-05, | |
| "loss": 2.7618, | |
| "step": 362000 | |
| }, | |
| { | |
| "epoch": 19.53, | |
| "learning_rate": 2.617141511655286e-05, | |
| "loss": 2.7633, | |
| "step": 363000 | |
| }, | |
| { | |
| "epoch": 19.58, | |
| "learning_rate": 2.323404756298564e-05, | |
| "loss": 2.7611, | |
| "step": 364000 | |
| }, | |
| { | |
| "epoch": 19.64, | |
| "learning_rate": 2.0290793501295033e-05, | |
| "loss": 2.7599, | |
| "step": 365000 | |
| }, | |
| { | |
| "epoch": 19.69, | |
| "learning_rate": 1.7347539439604427e-05, | |
| "loss": 2.7633, | |
| "step": 366000 | |
| }, | |
| { | |
| "epoch": 19.74, | |
| "learning_rate": 1.4404285377913822e-05, | |
| "loss": 2.7611, | |
| "step": 367000 | |
| }, | |
| { | |
| "epoch": 19.8, | |
| "learning_rate": 1.1463974570284906e-05, | |
| "loss": 2.7643, | |
| "step": 368000 | |
| }, | |
| { | |
| "epoch": 19.85, | |
| "learning_rate": 8.520720508594302e-06, | |
| "loss": 2.7627, | |
| "step": 369000 | |
| }, | |
| { | |
| "epoch": 19.91, | |
| "learning_rate": 5.580409700965387e-06, | |
| "loss": 2.7614, | |
| "step": 370000 | |
| }, | |
| { | |
| "epoch": 19.96, | |
| "learning_rate": 2.6371556392747826e-06, | |
| "loss": 2.7635, | |
| "step": 371000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.4103301921111753, | |
| "eval_loss": 3.405571460723877, | |
| "eval_runtime": 153.5852, | |
| "eval_samples_per_second": 377.113, | |
| "eval_steps_per_second": 5.892, | |
| "step": 371760 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 371760, | |
| "total_flos": 1.5663200871168e+18, | |
| "train_loss": 3.0311791784665343, | |
| "train_runtime": 80855.4689, | |
| "train_samples_per_second": 147.128, | |
| "train_steps_per_second": 4.598 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 371760, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 5000, | |
| "total_flos": 1.5663200871168e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |