| { | |
| "best_metric": 0.19522710144519806, | |
| "best_model_checkpoint": "./results_train/roberta-base/sst2/checkpoint-3500", | |
| "epoch": 10.0, | |
| "global_step": 42100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 3.9588281868566905e-06, | |
| "loss": 0.575, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_accuracy": 0.9071100917431193, | |
| "eval_loss": 0.2664913535118103, | |
| "eval_runtime": 2.4433, | |
| "eval_samples_per_second": 356.901, | |
| "eval_steps_per_second": 44.613, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 7.917656373713381e-06, | |
| "loss": 0.2989, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_accuracy": 0.9220183486238532, | |
| "eval_loss": 0.20883557200431824, | |
| "eval_runtime": 2.4454, | |
| "eval_samples_per_second": 356.584, | |
| "eval_steps_per_second": 44.573, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 1.1876484560570072e-05, | |
| "loss": 0.2725, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_accuracy": 0.9243119266055045, | |
| "eval_loss": 0.25596883893013, | |
| "eval_runtime": 2.451, | |
| "eval_samples_per_second": 355.775, | |
| "eval_steps_per_second": 44.472, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 1.5835312747426762e-05, | |
| "loss": 0.2814, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_accuracy": 0.926605504587156, | |
| "eval_loss": 0.20158442854881287, | |
| "eval_runtime": 2.462, | |
| "eval_samples_per_second": 354.188, | |
| "eval_steps_per_second": 44.274, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 1.9794140934283453e-05, | |
| "loss": 0.2586, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "eval_accuracy": 0.9174311926605505, | |
| "eval_loss": 0.22930225729942322, | |
| "eval_runtime": 2.4517, | |
| "eval_samples_per_second": 355.671, | |
| "eval_steps_per_second": 44.459, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.9760448779501697e-05, | |
| "loss": 0.2536, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_accuracy": 0.9323394495412844, | |
| "eval_loss": 0.23396578431129456, | |
| "eval_runtime": 2.4584, | |
| "eval_samples_per_second": 354.697, | |
| "eval_steps_per_second": 44.337, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 1.95077576186385e-05, | |
| "loss": 0.2494, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "eval_accuracy": 0.9323394495412844, | |
| "eval_loss": 0.19522710144519806, | |
| "eval_runtime": 2.4521, | |
| "eval_samples_per_second": 355.616, | |
| "eval_steps_per_second": 44.452, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 1.925506645777531e-05, | |
| "loss": 0.2396, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_accuracy": 0.9323394495412844, | |
| "eval_loss": 0.24936608970165253, | |
| "eval_runtime": 2.4569, | |
| "eval_samples_per_second": 354.916, | |
| "eval_steps_per_second": 44.365, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 1.9002375296912114e-05, | |
| "loss": 0.2123, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_accuracy": 0.9380733944954128, | |
| "eval_loss": 0.21870844066143036, | |
| "eval_runtime": 2.449, | |
| "eval_samples_per_second": 356.068, | |
| "eval_steps_per_second": 44.509, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 1.874968413604892e-05, | |
| "loss": 0.2042, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "eval_accuracy": 0.9151376146788991, | |
| "eval_loss": 0.2811821401119232, | |
| "eval_runtime": 2.4602, | |
| "eval_samples_per_second": 354.439, | |
| "eval_steps_per_second": 44.305, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 1.849699297518573e-05, | |
| "loss": 0.2083, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_accuracy": 0.9346330275229358, | |
| "eval_loss": 0.27386215329170227, | |
| "eval_runtime": 2.5255, | |
| "eval_samples_per_second": 345.272, | |
| "eval_steps_per_second": 43.159, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 1.8244301814322537e-05, | |
| "loss": 0.2041, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "eval_accuracy": 0.9380733944954128, | |
| "eval_loss": 0.20871196687221527, | |
| "eval_runtime": 2.4547, | |
| "eval_samples_per_second": 355.241, | |
| "eval_steps_per_second": 44.405, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 1.7991610653459345e-05, | |
| "loss": 0.1969, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "eval_accuracy": 0.9254587155963303, | |
| "eval_loss": 0.25904807448387146, | |
| "eval_runtime": 2.4532, | |
| "eval_samples_per_second": 355.448, | |
| "eval_steps_per_second": 44.431, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 1.773891949259615e-05, | |
| "loss": 0.1982, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "eval_accuracy": 0.930045871559633, | |
| "eval_loss": 0.2444588840007782, | |
| "eval_runtime": 2.4545, | |
| "eval_samples_per_second": 355.268, | |
| "eval_steps_per_second": 44.409, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 1.7486228331732958e-05, | |
| "loss": 0.1943, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "eval_accuracy": 0.926605504587156, | |
| "eval_loss": 0.2798321545124054, | |
| "eval_runtime": 2.4455, | |
| "eval_samples_per_second": 356.567, | |
| "eval_steps_per_second": 44.571, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 1.7233537170869766e-05, | |
| "loss": 0.1848, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "eval_accuracy": 0.9311926605504587, | |
| "eval_loss": 0.2844010591506958, | |
| "eval_runtime": 2.4586, | |
| "eval_samples_per_second": 354.679, | |
| "eval_steps_per_second": 44.335, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 1.698084601000657e-05, | |
| "loss": 0.1788, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "eval_accuracy": 0.9254587155963303, | |
| "eval_loss": 0.2998378872871399, | |
| "eval_runtime": 2.446, | |
| "eval_samples_per_second": 356.496, | |
| "eval_steps_per_second": 44.562, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 1.672815484914338e-05, | |
| "loss": 0.1623, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "eval_accuracy": 0.9392201834862385, | |
| "eval_loss": 0.2695905268192291, | |
| "eval_runtime": 2.4607, | |
| "eval_samples_per_second": 354.365, | |
| "eval_steps_per_second": 44.296, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 1.6475463688280183e-05, | |
| "loss": 0.1499, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "eval_accuracy": 0.9277522935779816, | |
| "eval_loss": 0.25331878662109375, | |
| "eval_runtime": 2.4449, | |
| "eval_samples_per_second": 356.659, | |
| "eval_steps_per_second": 44.582, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 1.622277252741699e-05, | |
| "loss": 0.1426, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "eval_accuracy": 0.930045871559633, | |
| "eval_loss": 0.29705262184143066, | |
| "eval_runtime": 2.4651, | |
| "eval_samples_per_second": 353.733, | |
| "eval_steps_per_second": 44.217, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 1.59700813665538e-05, | |
| "loss": 0.1479, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "eval_accuracy": 0.9357798165137615, | |
| "eval_loss": 0.25958266854286194, | |
| "eval_runtime": 2.4502, | |
| "eval_samples_per_second": 355.883, | |
| "eval_steps_per_second": 44.485, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 1.5717390205690607e-05, | |
| "loss": 0.1405, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "eval_accuracy": 0.9254587155963303, | |
| "eval_loss": 0.2944609522819519, | |
| "eval_runtime": 2.4554, | |
| "eval_samples_per_second": 355.141, | |
| "eval_steps_per_second": 44.393, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 1.5464699044827415e-05, | |
| "loss": 0.1577, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "eval_accuracy": 0.9002293577981652, | |
| "eval_loss": 0.40612396597862244, | |
| "eval_runtime": 2.4539, | |
| "eval_samples_per_second": 355.36, | |
| "eval_steps_per_second": 44.42, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 1.521200788396422e-05, | |
| "loss": 0.1521, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "eval_accuracy": 0.9334862385321101, | |
| "eval_loss": 0.2724354565143585, | |
| "eval_runtime": 2.4461, | |
| "eval_samples_per_second": 356.483, | |
| "eval_steps_per_second": 44.56, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 1.4959316723101027e-05, | |
| "loss": 0.1426, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "eval_accuracy": 0.9426605504587156, | |
| "eval_loss": 0.27123740315437317, | |
| "eval_runtime": 2.4449, | |
| "eval_samples_per_second": 356.655, | |
| "eval_steps_per_second": 44.582, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 1.4706625562237835e-05, | |
| "loss": 0.1206, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "eval_accuracy": 0.9357798165137615, | |
| "eval_loss": 0.2954227328300476, | |
| "eval_runtime": 2.467, | |
| "eval_samples_per_second": 353.464, | |
| "eval_steps_per_second": 44.183, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 1.4453934401374641e-05, | |
| "loss": 0.1074, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "eval_accuracy": 0.9392201834862385, | |
| "eval_loss": 0.2653304934501648, | |
| "eval_runtime": 2.4486, | |
| "eval_samples_per_second": 356.118, | |
| "eval_steps_per_second": 44.515, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 1.420124324051145e-05, | |
| "loss": 0.112, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "eval_accuracy": 0.9346330275229358, | |
| "eval_loss": 0.2777578830718994, | |
| "eval_runtime": 2.4566, | |
| "eval_samples_per_second": 354.969, | |
| "eval_steps_per_second": 44.371, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 1.3948552079648254e-05, | |
| "loss": 0.1147, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "eval_accuracy": 0.9311926605504587, | |
| "eval_loss": 0.3704558312892914, | |
| "eval_runtime": 2.4454, | |
| "eval_samples_per_second": 356.589, | |
| "eval_steps_per_second": 44.574, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 1.3695860918785062e-05, | |
| "loss": 0.1196, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "eval_accuracy": 0.9346330275229358, | |
| "eval_loss": 0.2889645993709564, | |
| "eval_runtime": 2.4563, | |
| "eval_samples_per_second": 354.999, | |
| "eval_steps_per_second": 44.375, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 1.344316975792187e-05, | |
| "loss": 0.1159, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "eval_accuracy": 0.926605504587156, | |
| "eval_loss": 0.3448694944381714, | |
| "eval_runtime": 2.4429, | |
| "eval_samples_per_second": 356.949, | |
| "eval_steps_per_second": 44.619, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 1.3190478597058676e-05, | |
| "loss": 0.119, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "eval_accuracy": 0.9334862385321101, | |
| "eval_loss": 0.3207152187824249, | |
| "eval_runtime": 2.461, | |
| "eval_samples_per_second": 354.323, | |
| "eval_steps_per_second": 44.29, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 1.2937787436195484e-05, | |
| "loss": 0.1268, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "eval_accuracy": 0.9311926605504587, | |
| "eval_loss": 0.3234628736972809, | |
| "eval_runtime": 2.4504, | |
| "eval_samples_per_second": 355.858, | |
| "eval_steps_per_second": 44.482, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 1.2685096275332289e-05, | |
| "loss": 0.1074, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "eval_accuracy": 0.9334862385321101, | |
| "eval_loss": 0.3650290369987488, | |
| "eval_runtime": 2.456, | |
| "eval_samples_per_second": 355.052, | |
| "eval_steps_per_second": 44.382, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "learning_rate": 1.2432405114469096e-05, | |
| "loss": 0.0805, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "eval_accuracy": 0.9380733944954128, | |
| "eval_loss": 0.33378419280052185, | |
| "eval_runtime": 2.4457, | |
| "eval_samples_per_second": 356.538, | |
| "eval_steps_per_second": 44.567, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 1.2179713953605903e-05, | |
| "loss": 0.0838, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "eval_accuracy": 0.9208715596330275, | |
| "eval_loss": 0.4302394688129425, | |
| "eval_runtime": 2.4587, | |
| "eval_samples_per_second": 354.661, | |
| "eval_steps_per_second": 44.333, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "learning_rate": 1.192702279274271e-05, | |
| "loss": 0.0848, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "eval_accuracy": 0.9323394495412844, | |
| "eval_loss": 0.40956971049308777, | |
| "eval_runtime": 2.4483, | |
| "eval_samples_per_second": 356.162, | |
| "eval_steps_per_second": 44.52, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "learning_rate": 1.1674331631879519e-05, | |
| "loss": 0.0922, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "eval_accuracy": 0.9369266055045872, | |
| "eval_loss": 0.3332035541534424, | |
| "eval_runtime": 2.4597, | |
| "eval_samples_per_second": 354.511, | |
| "eval_steps_per_second": 44.314, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "learning_rate": 1.1421640471016325e-05, | |
| "loss": 0.091, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "eval_accuracy": 0.9438073394495413, | |
| "eval_loss": 0.3024330735206604, | |
| "eval_runtime": 2.4457, | |
| "eval_samples_per_second": 356.542, | |
| "eval_steps_per_second": 44.568, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "learning_rate": 1.1168949310153133e-05, | |
| "loss": 0.0977, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "eval_accuracy": 0.9495412844036697, | |
| "eval_loss": 0.2673788070678711, | |
| "eval_runtime": 2.4587, | |
| "eval_samples_per_second": 354.654, | |
| "eval_steps_per_second": 44.332, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 1.0916258149289937e-05, | |
| "loss": 0.0897, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "eval_accuracy": 0.930045871559633, | |
| "eval_loss": 0.39930590987205505, | |
| "eval_runtime": 2.4473, | |
| "eval_samples_per_second": 356.313, | |
| "eval_steps_per_second": 44.539, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "learning_rate": 1.0663566988426745e-05, | |
| "loss": 0.1013, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "eval_accuracy": 0.9288990825688074, | |
| "eval_loss": 0.322666198015213, | |
| "eval_runtime": 2.4496, | |
| "eval_samples_per_second": 355.981, | |
| "eval_steps_per_second": 44.498, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 1.0410875827563553e-05, | |
| "loss": 0.0671, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "eval_accuracy": 0.9426605504587156, | |
| "eval_loss": 0.3374435603618622, | |
| "eval_runtime": 2.4457, | |
| "eval_samples_per_second": 356.54, | |
| "eval_steps_per_second": 44.567, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 5.23, | |
| "learning_rate": 1.015818466670036e-05, | |
| "loss": 0.0671, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 5.23, | |
| "eval_accuracy": 0.9277522935779816, | |
| "eval_loss": 0.4108366072177887, | |
| "eval_runtime": 2.4551, | |
| "eval_samples_per_second": 355.179, | |
| "eval_steps_per_second": 44.397, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 5.34, | |
| "learning_rate": 9.905493505837167e-06, | |
| "loss": 0.0652, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 5.34, | |
| "eval_accuracy": 0.9380733944954128, | |
| "eval_loss": 0.3549734652042389, | |
| "eval_runtime": 2.4475, | |
| "eval_samples_per_second": 356.289, | |
| "eval_steps_per_second": 44.536, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "learning_rate": 9.652802344973974e-06, | |
| "loss": 0.0664, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "eval_accuracy": 0.9357798165137615, | |
| "eval_loss": 0.339821994304657, | |
| "eval_runtime": 2.4559, | |
| "eval_samples_per_second": 355.062, | |
| "eval_steps_per_second": 44.383, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 5.58, | |
| "learning_rate": 9.40011118411078e-06, | |
| "loss": 0.0742, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 5.58, | |
| "eval_accuracy": 0.9380733944954128, | |
| "eval_loss": 0.3286002278327942, | |
| "eval_runtime": 2.4471, | |
| "eval_samples_per_second": 356.342, | |
| "eval_steps_per_second": 44.543, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "learning_rate": 9.147420023247588e-06, | |
| "loss": 0.0758, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "eval_accuracy": 0.9311926605504587, | |
| "eval_loss": 0.32764118909835815, | |
| "eval_runtime": 2.4639, | |
| "eval_samples_per_second": 353.904, | |
| "eval_steps_per_second": 44.238, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 5.82, | |
| "learning_rate": 8.894728862384394e-06, | |
| "loss": 0.075, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 5.82, | |
| "eval_accuracy": 0.9369266055045872, | |
| "eval_loss": 0.32022935152053833, | |
| "eval_runtime": 2.4503, | |
| "eval_samples_per_second": 355.874, | |
| "eval_steps_per_second": 44.484, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "learning_rate": 8.642037701521202e-06, | |
| "loss": 0.0686, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "eval_accuracy": 0.9415137614678899, | |
| "eval_loss": 0.3481292426586151, | |
| "eval_runtime": 2.4555, | |
| "eval_samples_per_second": 355.12, | |
| "eval_steps_per_second": 44.39, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 8.389346540658008e-06, | |
| "loss": 0.0729, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "eval_accuracy": 0.9334862385321101, | |
| "eval_loss": 0.38161903619766235, | |
| "eval_runtime": 2.4476, | |
| "eval_samples_per_second": 356.27, | |
| "eval_steps_per_second": 44.534, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 6.18, | |
| "learning_rate": 8.136655379794816e-06, | |
| "loss": 0.0568, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 6.18, | |
| "eval_accuracy": 0.9380733944954128, | |
| "eval_loss": 0.31324318051338196, | |
| "eval_runtime": 2.4707, | |
| "eval_samples_per_second": 352.935, | |
| "eval_steps_per_second": 44.117, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "learning_rate": 7.883964218931623e-06, | |
| "loss": 0.0529, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "eval_accuracy": 0.930045871559633, | |
| "eval_loss": 0.3756808340549469, | |
| "eval_runtime": 2.4544, | |
| "eval_samples_per_second": 355.287, | |
| "eval_steps_per_second": 44.411, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 6.41, | |
| "learning_rate": 7.631273058068429e-06, | |
| "loss": 0.0506, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 6.41, | |
| "eval_accuracy": 0.9380733944954128, | |
| "eval_loss": 0.33958113193511963, | |
| "eval_runtime": 2.4531, | |
| "eval_samples_per_second": 355.471, | |
| "eval_steps_per_second": 44.434, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 6.53, | |
| "learning_rate": 7.378581897205236e-06, | |
| "loss": 0.0476, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 6.53, | |
| "eval_accuracy": 0.9403669724770642, | |
| "eval_loss": 0.3641544580459595, | |
| "eval_runtime": 2.4417, | |
| "eval_samples_per_second": 357.132, | |
| "eval_steps_per_second": 44.641, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 6.65, | |
| "learning_rate": 7.125890736342044e-06, | |
| "loss": 0.0555, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 6.65, | |
| "eval_accuracy": 0.9403669724770642, | |
| "eval_loss": 0.34298017621040344, | |
| "eval_runtime": 2.4463, | |
| "eval_samples_per_second": 356.452, | |
| "eval_steps_per_second": 44.556, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 6.77, | |
| "learning_rate": 6.87319957547885e-06, | |
| "loss": 0.0574, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 6.77, | |
| "eval_accuracy": 0.9392201834862385, | |
| "eval_loss": 0.3401435613632202, | |
| "eval_runtime": 2.4439, | |
| "eval_samples_per_second": 356.811, | |
| "eval_steps_per_second": 44.601, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 6.89, | |
| "learning_rate": 6.620508414615657e-06, | |
| "loss": 0.0524, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 6.89, | |
| "eval_accuracy": 0.9346330275229358, | |
| "eval_loss": 0.33783158659935, | |
| "eval_runtime": 2.4521, | |
| "eval_samples_per_second": 355.616, | |
| "eval_steps_per_second": 44.452, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 6.367817253752464e-06, | |
| "loss": 0.0492, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "eval_accuracy": 0.9380733944954128, | |
| "eval_loss": 0.3833492398262024, | |
| "eval_runtime": 2.4457, | |
| "eval_samples_per_second": 356.538, | |
| "eval_steps_per_second": 44.567, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 7.13, | |
| "learning_rate": 6.1151260928892706e-06, | |
| "loss": 0.039, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 7.13, | |
| "eval_accuracy": 0.9346330275229358, | |
| "eval_loss": 0.3346712589263916, | |
| "eval_runtime": 2.4434, | |
| "eval_samples_per_second": 356.873, | |
| "eval_steps_per_second": 44.609, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 7.24, | |
| "learning_rate": 5.8624349320260785e-06, | |
| "loss": 0.0411, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 7.24, | |
| "eval_accuracy": 0.9334862385321101, | |
| "eval_loss": 0.4404141902923584, | |
| "eval_runtime": 2.4419, | |
| "eval_samples_per_second": 357.102, | |
| "eval_steps_per_second": 44.638, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 7.36, | |
| "learning_rate": 5.609743771162886e-06, | |
| "loss": 0.0412, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 7.36, | |
| "eval_accuracy": 0.9380733944954128, | |
| "eval_loss": 0.36179476976394653, | |
| "eval_runtime": 2.4414, | |
| "eval_samples_per_second": 357.173, | |
| "eval_steps_per_second": 44.647, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "learning_rate": 5.357052610299692e-06, | |
| "loss": 0.0477, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "eval_accuracy": 0.9380733944954128, | |
| "eval_loss": 0.3806387484073639, | |
| "eval_runtime": 2.4471, | |
| "eval_samples_per_second": 356.337, | |
| "eval_steps_per_second": 44.542, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "learning_rate": 5.104361449436499e-06, | |
| "loss": 0.0435, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "eval_accuracy": 0.9334862385321101, | |
| "eval_loss": 0.39115917682647705, | |
| "eval_runtime": 2.4665, | |
| "eval_samples_per_second": 353.536, | |
| "eval_steps_per_second": 44.192, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "learning_rate": 4.851670288573306e-06, | |
| "loss": 0.0443, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "eval_accuracy": 0.9392201834862385, | |
| "eval_loss": 0.39003145694732666, | |
| "eval_runtime": 2.4534, | |
| "eval_samples_per_second": 355.426, | |
| "eval_steps_per_second": 44.428, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 7.84, | |
| "learning_rate": 4.598979127710113e-06, | |
| "loss": 0.0421, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 7.84, | |
| "eval_accuracy": 0.9369266055045872, | |
| "eval_loss": 0.4152164161205292, | |
| "eval_runtime": 2.4525, | |
| "eval_samples_per_second": 355.556, | |
| "eval_steps_per_second": 44.445, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 7.96, | |
| "learning_rate": 4.34628796684692e-06, | |
| "loss": 0.0495, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 7.96, | |
| "eval_accuracy": 0.9288990825688074, | |
| "eval_loss": 0.3831779360771179, | |
| "eval_runtime": 2.447, | |
| "eval_samples_per_second": 356.361, | |
| "eval_steps_per_second": 44.545, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 8.08, | |
| "learning_rate": 4.093596805983727e-06, | |
| "loss": 0.0293, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 8.08, | |
| "eval_accuracy": 0.9346330275229358, | |
| "eval_loss": 0.44268128275871277, | |
| "eval_runtime": 2.4587, | |
| "eval_samples_per_second": 354.661, | |
| "eval_steps_per_second": 44.333, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 8.19, | |
| "learning_rate": 3.840905645120534e-06, | |
| "loss": 0.0253, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 8.19, | |
| "eval_accuracy": 0.9380733944954128, | |
| "eval_loss": 0.44246163964271545, | |
| "eval_runtime": 2.4427, | |
| "eval_samples_per_second": 356.983, | |
| "eval_steps_per_second": 44.623, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 8.31, | |
| "learning_rate": 3.5882144842573407e-06, | |
| "loss": 0.0407, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 8.31, | |
| "eval_accuracy": 0.9357798165137615, | |
| "eval_loss": 0.41019341349601746, | |
| "eval_runtime": 2.453, | |
| "eval_samples_per_second": 355.477, | |
| "eval_steps_per_second": 44.435, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 8.43, | |
| "learning_rate": 3.3355233233941482e-06, | |
| "loss": 0.0311, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 8.43, | |
| "eval_accuracy": 0.9369266055045872, | |
| "eval_loss": 0.44467687606811523, | |
| "eval_runtime": 2.4425, | |
| "eval_samples_per_second": 357.013, | |
| "eval_steps_per_second": 44.627, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "learning_rate": 3.082832162530955e-06, | |
| "loss": 0.0291, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "eval_accuracy": 0.9346330275229358, | |
| "eval_loss": 0.46120545268058777, | |
| "eval_runtime": 2.4514, | |
| "eval_samples_per_second": 355.714, | |
| "eval_steps_per_second": 44.464, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 8.67, | |
| "learning_rate": 2.8301410016677616e-06, | |
| "loss": 0.035, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 8.67, | |
| "eval_accuracy": 0.9346330275229358, | |
| "eval_loss": 0.4240852892398834, | |
| "eval_runtime": 2.4477, | |
| "eval_samples_per_second": 356.249, | |
| "eval_steps_per_second": 44.531, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 8.79, | |
| "learning_rate": 2.577449840804569e-06, | |
| "loss": 0.0381, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 8.79, | |
| "eval_accuracy": 0.9311926605504587, | |
| "eval_loss": 0.41976186633110046, | |
| "eval_runtime": 2.4523, | |
| "eval_samples_per_second": 355.586, | |
| "eval_steps_per_second": 44.448, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 8.91, | |
| "learning_rate": 2.3247586799413758e-06, | |
| "loss": 0.0234, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 8.91, | |
| "eval_accuracy": 0.9369266055045872, | |
| "eval_loss": 0.4344768822193146, | |
| "eval_runtime": 2.4469, | |
| "eval_samples_per_second": 356.366, | |
| "eval_steps_per_second": 44.546, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 2.072067519078183e-06, | |
| "loss": 0.0311, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "eval_accuracy": 0.9311926605504587, | |
| "eval_loss": 0.45580777525901794, | |
| "eval_runtime": 2.4545, | |
| "eval_samples_per_second": 355.27, | |
| "eval_steps_per_second": 44.409, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 9.14, | |
| "learning_rate": 1.8193763582149898e-06, | |
| "loss": 0.028, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 9.14, | |
| "eval_accuracy": 0.9380733944954128, | |
| "eval_loss": 0.42450448870658875, | |
| "eval_runtime": 2.4449, | |
| "eval_samples_per_second": 356.658, | |
| "eval_steps_per_second": 44.582, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 9.26, | |
| "learning_rate": 1.5666851973517969e-06, | |
| "loss": 0.0213, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 9.26, | |
| "eval_accuracy": 0.9380733944954128, | |
| "eval_loss": 0.446162611246109, | |
| "eval_runtime": 2.4606, | |
| "eval_samples_per_second": 354.384, | |
| "eval_steps_per_second": 44.298, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "learning_rate": 1.3139940364886035e-06, | |
| "loss": 0.0276, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "eval_accuracy": 0.9380733944954128, | |
| "eval_loss": 0.42100322246551514, | |
| "eval_runtime": 2.4512, | |
| "eval_samples_per_second": 355.743, | |
| "eval_steps_per_second": 44.468, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "learning_rate": 1.0613028756254106e-06, | |
| "loss": 0.0183, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "eval_accuracy": 0.9403669724770642, | |
| "eval_loss": 0.43098002672195435, | |
| "eval_runtime": 2.45, | |
| "eval_samples_per_second": 355.922, | |
| "eval_steps_per_second": 44.49, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 9.62, | |
| "learning_rate": 8.086117147622177e-07, | |
| "loss": 0.0184, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 9.62, | |
| "eval_accuracy": 0.9403669724770642, | |
| "eval_loss": 0.4437469244003296, | |
| "eval_runtime": 2.4461, | |
| "eval_samples_per_second": 356.492, | |
| "eval_steps_per_second": 44.561, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 9.74, | |
| "learning_rate": 5.559205538990246e-07, | |
| "loss": 0.0296, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 9.74, | |
| "eval_accuracy": 0.9392201834862385, | |
| "eval_loss": 0.43114030361175537, | |
| "eval_runtime": 2.4504, | |
| "eval_samples_per_second": 355.859, | |
| "eval_steps_per_second": 44.482, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 9.86, | |
| "learning_rate": 3.0322939303583163e-07, | |
| "loss": 0.019, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 9.86, | |
| "eval_accuracy": 0.9415137614678899, | |
| "eval_loss": 0.42435380816459656, | |
| "eval_runtime": 2.4473, | |
| "eval_samples_per_second": 356.311, | |
| "eval_steps_per_second": 44.539, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 9.98, | |
| "learning_rate": 5.053823217263861e-08, | |
| "loss": 0.0245, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 9.98, | |
| "eval_accuracy": 0.9415137614678899, | |
| "eval_loss": 0.42697247862815857, | |
| "eval_runtime": 2.46, | |
| "eval_samples_per_second": 354.474, | |
| "eval_steps_per_second": 44.309, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 42100, | |
| "total_flos": 4.43006661686016e+16, | |
| "train_loss": 0.10745611605338416, | |
| "train_runtime": 8358.8854, | |
| "train_samples_per_second": 80.572, | |
| "train_steps_per_second": 5.037 | |
| } | |
| ], | |
| "max_steps": 42100, | |
| "num_train_epochs": 10, | |
| "total_flos": 4.43006661686016e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |