| { | |
| "best_metric": 0.23302823305130005, | |
| "best_model_checkpoint": "./results_train/roberta-base/qnli/checkpoint-7000", | |
| "epoch": 10.0, | |
| "global_step": 65470, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 2.545176889793841e-06, | |
| "loss": 0.6919, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_accuracy": 0.7501372872048325, | |
| "eval_loss": 0.6161059141159058, | |
| "eval_runtime": 15.4359, | |
| "eval_samples_per_second": 353.916, | |
| "eval_steps_per_second": 44.248, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 5.090353779587682e-06, | |
| "loss": 0.4801, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_accuracy": 0.8550247116968699, | |
| "eval_loss": 0.3524323105812073, | |
| "eval_runtime": 15.4744, | |
| "eval_samples_per_second": 353.035, | |
| "eval_steps_per_second": 44.137, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 7.635530669381522e-06, | |
| "loss": 0.4049, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "eval_accuracy": 0.8742449203734212, | |
| "eval_loss": 0.301090806722641, | |
| "eval_runtime": 15.4636, | |
| "eval_samples_per_second": 353.282, | |
| "eval_steps_per_second": 44.168, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 1.0180707559175364e-05, | |
| "loss": 0.3827, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_accuracy": 0.876807614863628, | |
| "eval_loss": 0.3125079274177551, | |
| "eval_runtime": 15.5809, | |
| "eval_samples_per_second": 350.621, | |
| "eval_steps_per_second": 43.836, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 1.2725884448969203e-05, | |
| "loss": 0.3445, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "eval_accuracy": 0.8923668314113125, | |
| "eval_loss": 0.29156529903411865, | |
| "eval_runtime": 15.4606, | |
| "eval_samples_per_second": 353.351, | |
| "eval_steps_per_second": 44.177, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 1.5271061338763045e-05, | |
| "loss": 0.3567, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "eval_accuracy": 0.8991396668497162, | |
| "eval_loss": 0.2662096619606018, | |
| "eval_runtime": 15.4956, | |
| "eval_samples_per_second": 352.551, | |
| "eval_steps_per_second": 44.077, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 1.7816238228556887e-05, | |
| "loss": 0.3422, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_accuracy": 0.8980413692110562, | |
| "eval_loss": 0.2656656801700592, | |
| "eval_runtime": 15.5011, | |
| "eval_samples_per_second": 352.426, | |
| "eval_steps_per_second": 44.061, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 1.997692595180449e-05, | |
| "loss": 0.3257, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_accuracy": 0.9020684605528098, | |
| "eval_loss": 0.2830497622489929, | |
| "eval_runtime": 15.5605, | |
| "eval_samples_per_second": 351.081, | |
| "eval_steps_per_second": 43.893, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 1.9814432654652997e-05, | |
| "loss": 0.3506, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "eval_accuracy": 0.9062786015010068, | |
| "eval_loss": 0.24342051148414612, | |
| "eval_runtime": 15.5612, | |
| "eval_samples_per_second": 351.065, | |
| "eval_steps_per_second": 43.891, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 1.9651939357501505e-05, | |
| "loss": 0.317, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_accuracy": 0.9051803038623467, | |
| "eval_loss": 0.24402067065238953, | |
| "eval_runtime": 15.5216, | |
| "eval_samples_per_second": 351.96, | |
| "eval_steps_per_second": 44.003, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 1.9489446060350013e-05, | |
| "loss": 0.3152, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_accuracy": 0.9015193117334798, | |
| "eval_loss": 0.2785772681236267, | |
| "eval_runtime": 15.4885, | |
| "eval_samples_per_second": 352.712, | |
| "eval_steps_per_second": 44.097, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 1.932695276319852e-05, | |
| "loss": 0.2966, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_accuracy": 0.9082921471718836, | |
| "eval_loss": 0.2599336504936218, | |
| "eval_runtime": 15.5494, | |
| "eval_samples_per_second": 351.332, | |
| "eval_steps_per_second": 43.924, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 1.916445946604703e-05, | |
| "loss": 0.298, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.9070107999267801, | |
| "eval_loss": 0.2616613209247589, | |
| "eval_runtime": 16.1579, | |
| "eval_samples_per_second": 338.101, | |
| "eval_steps_per_second": 42.27, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 1.9001966168895533e-05, | |
| "loss": 0.2634, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_accuracy": 0.9154310818231741, | |
| "eval_loss": 0.23302823305130005, | |
| "eval_runtime": 15.5105, | |
| "eval_samples_per_second": 352.213, | |
| "eval_steps_per_second": 44.035, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 1.883947287174404e-05, | |
| "loss": 0.2625, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "eval_accuracy": 0.9108548416620904, | |
| "eval_loss": 0.259826123714447, | |
| "eval_runtime": 16.1363, | |
| "eval_samples_per_second": 338.553, | |
| "eval_steps_per_second": 42.327, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 1.867697957459255e-05, | |
| "loss": 0.2596, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "eval_accuracy": 0.9099395936298736, | |
| "eval_loss": 0.36161935329437256, | |
| "eval_runtime": 15.5589, | |
| "eval_samples_per_second": 351.116, | |
| "eval_steps_per_second": 43.898, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 1.8514486277441056e-05, | |
| "loss": 0.2457, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_accuracy": 0.909573494416987, | |
| "eval_loss": 0.2799856662750244, | |
| "eval_runtime": 15.5157, | |
| "eval_samples_per_second": 352.095, | |
| "eval_steps_per_second": 44.02, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 1.8351992980289564e-05, | |
| "loss": 0.2545, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "eval_accuracy": 0.9081090975654402, | |
| "eval_loss": 0.29598304629325867, | |
| "eval_runtime": 15.4941, | |
| "eval_samples_per_second": 352.587, | |
| "eval_steps_per_second": 44.081, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 1.8189499683138072e-05, | |
| "loss": 0.2535, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "eval_accuracy": 0.9114039904814205, | |
| "eval_loss": 0.23894734680652618, | |
| "eval_runtime": 16.236, | |
| "eval_samples_per_second": 336.474, | |
| "eval_steps_per_second": 42.067, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 1.802700638598658e-05, | |
| "loss": 0.2639, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_accuracy": 0.8912685337726524, | |
| "eval_loss": 0.33427268266677856, | |
| "eval_runtime": 15.4977, | |
| "eval_samples_per_second": 352.503, | |
| "eval_steps_per_second": 44.071, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 1.7864513088835088e-05, | |
| "loss": 0.2434, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_accuracy": 0.9115870400878638, | |
| "eval_loss": 0.24697424471378326, | |
| "eval_runtime": 15.5007, | |
| "eval_samples_per_second": 352.436, | |
| "eval_steps_per_second": 44.063, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 1.7702019791683592e-05, | |
| "loss": 0.2613, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_accuracy": 0.9092073952041003, | |
| "eval_loss": 0.294897198677063, | |
| "eval_runtime": 15.4616, | |
| "eval_samples_per_second": 353.327, | |
| "eval_steps_per_second": 44.174, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 1.75395264945321e-05, | |
| "loss": 0.2456, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "eval_accuracy": 0.9163463298553908, | |
| "eval_loss": 0.2556602358818054, | |
| "eval_runtime": 15.4884, | |
| "eval_samples_per_second": 352.715, | |
| "eval_steps_per_second": 44.097, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 1.737703319738061e-05, | |
| "loss": 0.2483, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "eval_accuracy": 0.9141497345780707, | |
| "eval_loss": 0.24623550474643707, | |
| "eval_runtime": 15.544, | |
| "eval_samples_per_second": 351.453, | |
| "eval_steps_per_second": 43.94, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 1.7214539900229116e-05, | |
| "loss": 0.2524, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "eval_accuracy": 0.9114039904814205, | |
| "eval_loss": 0.2453071027994156, | |
| "eval_runtime": 15.5138, | |
| "eval_samples_per_second": 352.139, | |
| "eval_steps_per_second": 44.025, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 1.7052046603077624e-05, | |
| "loss": 0.2467, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_accuracy": 0.9161632802489474, | |
| "eval_loss": 0.2610779404640198, | |
| "eval_runtime": 15.5605, | |
| "eval_samples_per_second": 351.082, | |
| "eval_steps_per_second": 43.893, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 1.6889553305926132e-05, | |
| "loss": 0.2059, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "eval_accuracy": 0.9157971810360608, | |
| "eval_loss": 0.3070930540561676, | |
| "eval_runtime": 15.4895, | |
| "eval_samples_per_second": 352.691, | |
| "eval_steps_per_second": 44.094, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 1.672706000877464e-05, | |
| "loss": 0.1968, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "eval_accuracy": 0.9209225700164745, | |
| "eval_loss": 0.3205060660839081, | |
| "eval_runtime": 15.5462, | |
| "eval_samples_per_second": 351.404, | |
| "eval_steps_per_second": 43.933, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 1.6564566711623148e-05, | |
| "loss": 0.1944, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "eval_accuracy": 0.9145158337909574, | |
| "eval_loss": 0.342997282743454, | |
| "eval_runtime": 15.462, | |
| "eval_samples_per_second": 353.317, | |
| "eval_steps_per_second": 44.173, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 1.6402073414471655e-05, | |
| "loss": 0.2065, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "eval_accuracy": 0.9146988833974007, | |
| "eval_loss": 0.33879798650741577, | |
| "eval_runtime": 15.4872, | |
| "eval_samples_per_second": 352.743, | |
| "eval_steps_per_second": 44.101, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 1.6239580117320163e-05, | |
| "loss": 0.1992, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "eval_accuracy": 0.9157971810360608, | |
| "eval_loss": 0.2569463551044464, | |
| "eval_runtime": 15.4551, | |
| "eval_samples_per_second": 353.476, | |
| "eval_steps_per_second": 44.193, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 1.607708682016867e-05, | |
| "loss": 0.1994, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "eval_accuracy": 0.9108548416620904, | |
| "eval_loss": 0.33493661880493164, | |
| "eval_runtime": 15.5323, | |
| "eval_samples_per_second": 351.718, | |
| "eval_steps_per_second": 43.973, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 1.5914593523017176e-05, | |
| "loss": 0.2001, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "eval_accuracy": 0.909573494416987, | |
| "eval_loss": 0.2850426137447357, | |
| "eval_runtime": 15.4842, | |
| "eval_samples_per_second": 352.811, | |
| "eval_steps_per_second": 44.109, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 1.5752100225865684e-05, | |
| "loss": 0.2014, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "eval_accuracy": 0.9200073219842577, | |
| "eval_loss": 0.3214400112628937, | |
| "eval_runtime": 15.5272, | |
| "eval_samples_per_second": 351.833, | |
| "eval_steps_per_second": 43.987, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 1.558960692871419e-05, | |
| "loss": 0.2156, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "eval_accuracy": 0.9134175361522973, | |
| "eval_loss": 0.3078743815422058, | |
| "eval_runtime": 15.5189, | |
| "eval_samples_per_second": 352.023, | |
| "eval_steps_per_second": 44.011, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 1.54271136315627e-05, | |
| "loss": 0.2036, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "eval_accuracy": 0.9163463298553908, | |
| "eval_loss": 0.2739076316356659, | |
| "eval_runtime": 15.5712, | |
| "eval_samples_per_second": 350.84, | |
| "eval_steps_per_second": 43.863, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 1.5264620334411207e-05, | |
| "loss": 0.2118, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "eval_accuracy": 0.918542925132711, | |
| "eval_loss": 0.2790161669254303, | |
| "eval_runtime": 15.4896, | |
| "eval_samples_per_second": 352.687, | |
| "eval_steps_per_second": 44.094, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 1.5102127037259715e-05, | |
| "loss": 0.2167, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "eval_accuracy": 0.9167124290682775, | |
| "eval_loss": 0.2698538899421692, | |
| "eval_runtime": 15.51, | |
| "eval_samples_per_second": 352.224, | |
| "eval_steps_per_second": 44.036, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 1.4939633740108221e-05, | |
| "loss": 0.2015, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "eval_accuracy": 0.9189090243455976, | |
| "eval_loss": 0.28947147727012634, | |
| "eval_runtime": 15.4881, | |
| "eval_samples_per_second": 352.721, | |
| "eval_steps_per_second": 44.098, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 1.477714044295673e-05, | |
| "loss": 0.1649, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "eval_accuracy": 0.9161632802489474, | |
| "eval_loss": 0.37189313769340515, | |
| "eval_runtime": 15.48, | |
| "eval_samples_per_second": 352.907, | |
| "eval_steps_per_second": 44.121, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 1.4614647145805237e-05, | |
| "loss": 0.1505, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "eval_accuracy": 0.9132344865458539, | |
| "eval_loss": 0.3699614107608795, | |
| "eval_runtime": 15.4613, | |
| "eval_samples_per_second": 353.334, | |
| "eval_steps_per_second": 44.175, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 1.4452153848653745e-05, | |
| "loss": 0.1509, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "eval_accuracy": 0.9156141314296175, | |
| "eval_loss": 0.3720746636390686, | |
| "eval_runtime": 15.4811, | |
| "eval_samples_per_second": 352.881, | |
| "eval_steps_per_second": 44.118, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "learning_rate": 1.4289660551502251e-05, | |
| "loss": 0.1517, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "eval_accuracy": 0.9154310818231741, | |
| "eval_loss": 0.35659804940223694, | |
| "eval_runtime": 15.4964, | |
| "eval_samples_per_second": 352.533, | |
| "eval_steps_per_second": 44.075, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 1.412716725435076e-05, | |
| "loss": 0.1583, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "eval_accuracy": 0.9139666849716274, | |
| "eval_loss": 0.3975317180156708, | |
| "eval_runtime": 15.5028, | |
| "eval_samples_per_second": 352.388, | |
| "eval_steps_per_second": 44.057, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "learning_rate": 1.3964673957199267e-05, | |
| "loss": 0.1568, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "eval_accuracy": 0.9136005857587406, | |
| "eval_loss": 0.4135216772556305, | |
| "eval_runtime": 15.5258, | |
| "eval_samples_per_second": 351.866, | |
| "eval_steps_per_second": 43.991, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 1.3802180660047775e-05, | |
| "loss": 0.1642, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "eval_accuracy": 0.9128683873329673, | |
| "eval_loss": 0.37052008509635925, | |
| "eval_runtime": 15.557, | |
| "eval_samples_per_second": 351.161, | |
| "eval_steps_per_second": 43.903, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 1.3639687362896281e-05, | |
| "loss": 0.1781, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "eval_accuracy": 0.9156141314296175, | |
| "eval_loss": 0.33989137411117554, | |
| "eval_runtime": 15.5332, | |
| "eval_samples_per_second": 351.699, | |
| "eval_steps_per_second": 43.97, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 1.347719406574479e-05, | |
| "loss": 0.1725, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "eval_accuracy": 0.9159802306425041, | |
| "eval_loss": 0.3164858818054199, | |
| "eval_runtime": 15.5466, | |
| "eval_samples_per_second": 351.394, | |
| "eval_steps_per_second": 43.932, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 1.3314700768593297e-05, | |
| "loss": 0.1675, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "eval_accuracy": 0.917993776313381, | |
| "eval_loss": 0.3278675973415375, | |
| "eval_runtime": 15.4883, | |
| "eval_samples_per_second": 352.717, | |
| "eval_steps_per_second": 44.098, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 1.3152207471441804e-05, | |
| "loss": 0.165, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "eval_accuracy": 0.920190371590701, | |
| "eval_loss": 0.34237906336784363, | |
| "eval_runtime": 15.4787, | |
| "eval_samples_per_second": 352.937, | |
| "eval_steps_per_second": 44.125, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "learning_rate": 1.298971417429031e-05, | |
| "loss": 0.1608, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 3.89, | |
| "eval_accuracy": 0.9137836353651839, | |
| "eval_loss": 0.4021676778793335, | |
| "eval_runtime": 15.432, | |
| "eval_samples_per_second": 354.004, | |
| "eval_steps_per_second": 44.259, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "learning_rate": 1.282722087713882e-05, | |
| "loss": 0.1576, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "eval_accuracy": 0.9146988833974007, | |
| "eval_loss": 0.3611091375350952, | |
| "eval_runtime": 15.4994, | |
| "eval_samples_per_second": 352.465, | |
| "eval_steps_per_second": 44.066, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 1.2664727579987326e-05, | |
| "loss": 0.1382, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "eval_accuracy": 0.9139666849716274, | |
| "eval_loss": 0.400115966796875, | |
| "eval_runtime": 15.4319, | |
| "eval_samples_per_second": 354.008, | |
| "eval_steps_per_second": 44.259, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 1.2502234282835834e-05, | |
| "loss": 0.1126, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "eval_accuracy": 0.9168954786747209, | |
| "eval_loss": 0.40148910880088806, | |
| "eval_runtime": 15.5197, | |
| "eval_samples_per_second": 352.005, | |
| "eval_steps_per_second": 44.009, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "learning_rate": 1.233974098568434e-05, | |
| "loss": 0.1048, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "eval_accuracy": 0.9168954786747209, | |
| "eval_loss": 0.391897052526474, | |
| "eval_runtime": 15.4767, | |
| "eval_samples_per_second": 352.982, | |
| "eval_steps_per_second": 44.131, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 1.217724768853285e-05, | |
| "loss": 0.1057, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "eval_accuracy": 0.9176276771004942, | |
| "eval_loss": 0.4071587324142456, | |
| "eval_runtime": 15.5017, | |
| "eval_samples_per_second": 352.414, | |
| "eval_steps_per_second": 44.06, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 1.2014754391381356e-05, | |
| "loss": 0.1212, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "eval_accuracy": 0.9161632802489474, | |
| "eval_loss": 0.36226457357406616, | |
| "eval_runtime": 15.4788, | |
| "eval_samples_per_second": 352.935, | |
| "eval_steps_per_second": 44.125, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "learning_rate": 1.1852261094229864e-05, | |
| "loss": 0.1152, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "eval_accuracy": 0.914881933003844, | |
| "eval_loss": 0.39458510279655457, | |
| "eval_runtime": 15.5093, | |
| "eval_samples_per_second": 352.239, | |
| "eval_steps_per_second": 44.038, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "learning_rate": 1.168976779707837e-05, | |
| "loss": 0.125, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "eval_accuracy": 0.9156141314296175, | |
| "eval_loss": 0.4142376780509949, | |
| "eval_runtime": 15.4899, | |
| "eval_samples_per_second": 352.682, | |
| "eval_steps_per_second": 44.093, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "learning_rate": 1.152727449992688e-05, | |
| "loss": 0.1195, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "eval_accuracy": 0.9150649826102873, | |
| "eval_loss": 0.40947625041007996, | |
| "eval_runtime": 15.5825, | |
| "eval_samples_per_second": 350.585, | |
| "eval_steps_per_second": 43.831, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 4.66, | |
| "learning_rate": 1.1364781202775386e-05, | |
| "loss": 0.1139, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 4.66, | |
| "eval_accuracy": 0.9088412959912137, | |
| "eval_loss": 0.4585929811000824, | |
| "eval_runtime": 15.5177, | |
| "eval_samples_per_second": 352.05, | |
| "eval_steps_per_second": 44.014, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "learning_rate": 1.1202287905623894e-05, | |
| "loss": 0.1279, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 4.73, | |
| "eval_accuracy": 0.9203734211971444, | |
| "eval_loss": 0.39000362157821655, | |
| "eval_runtime": 15.5995, | |
| "eval_samples_per_second": 350.204, | |
| "eval_steps_per_second": 43.784, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "learning_rate": 1.10397946084724e-05, | |
| "loss": 0.1306, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "eval_accuracy": 0.9165293794618341, | |
| "eval_loss": 0.37414777278900146, | |
| "eval_runtime": 15.5095, | |
| "eval_samples_per_second": 352.236, | |
| "eval_steps_per_second": 44.038, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 1.087730131132091e-05, | |
| "loss": 0.1091, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "eval_accuracy": 0.9207395204100312, | |
| "eval_loss": 0.4295918941497803, | |
| "eval_runtime": 15.4949, | |
| "eval_samples_per_second": 352.567, | |
| "eval_steps_per_second": 44.079, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "learning_rate": 1.0714808014169416e-05, | |
| "loss": 0.1272, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "eval_accuracy": 0.9189090243455976, | |
| "eval_loss": 0.37242451310157776, | |
| "eval_runtime": 15.5005, | |
| "eval_samples_per_second": 352.44, | |
| "eval_steps_per_second": 44.063, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 1.0552314717017924e-05, | |
| "loss": 0.0906, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "eval_accuracy": 0.9181768259198243, | |
| "eval_loss": 0.4511684477329254, | |
| "eval_runtime": 15.5205, | |
| "eval_samples_per_second": 351.985, | |
| "eval_steps_per_second": 44.006, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "learning_rate": 1.038982141986643e-05, | |
| "loss": 0.0915, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 5.12, | |
| "eval_accuracy": 0.9220208676551346, | |
| "eval_loss": 0.4160422086715698, | |
| "eval_runtime": 15.4765, | |
| "eval_samples_per_second": 352.987, | |
| "eval_steps_per_second": 44.131, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "learning_rate": 1.022732812271494e-05, | |
| "loss": 0.0773, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 5.19, | |
| "eval_accuracy": 0.917993776313381, | |
| "eval_loss": 0.47428572177886963, | |
| "eval_runtime": 15.4784, | |
| "eval_samples_per_second": 352.944, | |
| "eval_steps_per_second": 44.126, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 5.27, | |
| "learning_rate": 1.0064834825563446e-05, | |
| "loss": 0.0861, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 5.27, | |
| "eval_accuracy": 0.9203734211971444, | |
| "eval_loss": 0.5023528337478638, | |
| "eval_runtime": 15.4707, | |
| "eval_samples_per_second": 353.119, | |
| "eval_steps_per_second": 44.148, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "learning_rate": 9.902341528411954e-06, | |
| "loss": 0.0729, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "eval_accuracy": 0.9203734211971444, | |
| "eval_loss": 0.42824575304985046, | |
| "eval_runtime": 15.5357, | |
| "eval_samples_per_second": 351.643, | |
| "eval_steps_per_second": 43.963, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 5.42, | |
| "learning_rate": 9.739848231260461e-06, | |
| "loss": 0.0901, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 5.42, | |
| "eval_accuracy": 0.9225700164744646, | |
| "eval_loss": 0.46121296286582947, | |
| "eval_runtime": 15.4855, | |
| "eval_samples_per_second": 352.782, | |
| "eval_steps_per_second": 44.106, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "learning_rate": 9.57735493410897e-06, | |
| "loss": 0.0856, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "eval_accuracy": 0.917993776313381, | |
| "eval_loss": 0.44952550530433655, | |
| "eval_runtime": 15.5112, | |
| "eval_samples_per_second": 352.196, | |
| "eval_steps_per_second": 44.033, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 5.58, | |
| "learning_rate": 9.414861636957477e-06, | |
| "loss": 0.0839, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 5.58, | |
| "eval_accuracy": 0.9205564708035878, | |
| "eval_loss": 0.45009082555770874, | |
| "eval_runtime": 15.4775, | |
| "eval_samples_per_second": 352.963, | |
| "eval_steps_per_second": 44.128, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "learning_rate": 9.252368339805983e-06, | |
| "loss": 0.0874, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "eval_accuracy": 0.9200073219842577, | |
| "eval_loss": 0.41364917159080505, | |
| "eval_runtime": 15.5201, | |
| "eval_samples_per_second": 351.994, | |
| "eval_steps_per_second": 44.007, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 5.73, | |
| "learning_rate": 9.089875042654491e-06, | |
| "loss": 0.0944, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 5.73, | |
| "eval_accuracy": 0.9165293794618341, | |
| "eval_loss": 0.46293067932128906, | |
| "eval_runtime": 15.474, | |
| "eval_samples_per_second": 353.043, | |
| "eval_steps_per_second": 44.138, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "learning_rate": 8.927381745502999e-06, | |
| "loss": 0.0874, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "eval_accuracy": 0.9159802306425041, | |
| "eval_loss": 0.4790218770503998, | |
| "eval_runtime": 15.4804, | |
| "eval_samples_per_second": 352.897, | |
| "eval_steps_per_second": 44.12, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "learning_rate": 8.764888448351507e-06, | |
| "loss": 0.0859, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "eval_accuracy": 0.9132344865458539, | |
| "eval_loss": 0.4725111722946167, | |
| "eval_runtime": 15.4939, | |
| "eval_samples_per_second": 352.59, | |
| "eval_steps_per_second": 44.082, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "learning_rate": 8.602395151200013e-06, | |
| "loss": 0.0808, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "eval_accuracy": 0.9161632802489474, | |
| "eval_loss": 0.4613119959831238, | |
| "eval_runtime": 15.5324, | |
| "eval_samples_per_second": 351.717, | |
| "eval_steps_per_second": 43.973, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 8.439901854048521e-06, | |
| "loss": 0.0723, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "eval_accuracy": 0.9194581731649277, | |
| "eval_loss": 0.4815793037414551, | |
| "eval_runtime": 15.4962, | |
| "eval_samples_per_second": 352.537, | |
| "eval_steps_per_second": 44.075, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "learning_rate": 8.277408556897029e-06, | |
| "loss": 0.0568, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "eval_accuracy": 0.9187259747391543, | |
| "eval_loss": 0.5257057547569275, | |
| "eval_runtime": 15.5592, | |
| "eval_samples_per_second": 351.112, | |
| "eval_steps_per_second": 43.897, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 6.19, | |
| "learning_rate": 8.114915259745537e-06, | |
| "loss": 0.0628, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 6.19, | |
| "eval_accuracy": 0.9194581731649277, | |
| "eval_loss": 0.4516303837299347, | |
| "eval_runtime": 15.5547, | |
| "eval_samples_per_second": 351.213, | |
| "eval_steps_per_second": 43.91, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 6.26, | |
| "learning_rate": 7.952421962594043e-06, | |
| "loss": 0.053, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 6.26, | |
| "eval_accuracy": 0.9187259747391543, | |
| "eval_loss": 0.4928816258907318, | |
| "eval_runtime": 15.5194, | |
| "eval_samples_per_second": 352.012, | |
| "eval_steps_per_second": 44.01, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "learning_rate": 7.789928665442551e-06, | |
| "loss": 0.0574, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "eval_accuracy": 0.919092073952041, | |
| "eval_loss": 0.48882895708084106, | |
| "eval_runtime": 15.4527, | |
| "eval_samples_per_second": 353.531, | |
| "eval_steps_per_second": 44.2, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 6.42, | |
| "learning_rate": 7.627435368291059e-06, | |
| "loss": 0.0717, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 6.42, | |
| "eval_accuracy": 0.9165293794618341, | |
| "eval_loss": 0.4769105911254883, | |
| "eval_runtime": 15.4715, | |
| "eval_samples_per_second": 353.101, | |
| "eval_steps_per_second": 44.146, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 6.49, | |
| "learning_rate": 7.464942071139566e-06, | |
| "loss": 0.0622, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 6.49, | |
| "eval_accuracy": 0.9183598755262676, | |
| "eval_loss": 0.5081596374511719, | |
| "eval_runtime": 15.4674, | |
| "eval_samples_per_second": 353.193, | |
| "eval_steps_per_second": 44.157, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 6.57, | |
| "learning_rate": 7.302448773988074e-06, | |
| "loss": 0.0593, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 6.57, | |
| "eval_accuracy": 0.9211056196229178, | |
| "eval_loss": 0.44598302245140076, | |
| "eval_runtime": 15.4988, | |
| "eval_samples_per_second": 352.48, | |
| "eval_steps_per_second": 44.068, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 6.64, | |
| "learning_rate": 7.139955476836581e-06, | |
| "loss": 0.0603, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 6.64, | |
| "eval_accuracy": 0.9205564708035878, | |
| "eval_loss": 0.43451356887817383, | |
| "eval_runtime": 15.4424, | |
| "eval_samples_per_second": 353.766, | |
| "eval_steps_per_second": 44.229, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "learning_rate": 6.9774621796850885e-06, | |
| "loss": 0.0659, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "eval_accuracy": 0.9189090243455976, | |
| "eval_loss": 0.4423438310623169, | |
| "eval_runtime": 15.4892, | |
| "eval_samples_per_second": 352.697, | |
| "eval_steps_per_second": 44.095, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "learning_rate": 6.8149688825335956e-06, | |
| "loss": 0.0629, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "eval_accuracy": 0.919092073952041, | |
| "eval_loss": 0.47710955142974854, | |
| "eval_runtime": 15.4296, | |
| "eval_samples_per_second": 354.061, | |
| "eval_steps_per_second": 44.266, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 6.87, | |
| "learning_rate": 6.6524755853821034e-06, | |
| "loss": 0.058, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 6.87, | |
| "eval_accuracy": 0.9227530660809079, | |
| "eval_loss": 0.4589206576347351, | |
| "eval_runtime": 15.5247, | |
| "eval_samples_per_second": 351.892, | |
| "eval_steps_per_second": 43.995, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 6.95, | |
| "learning_rate": 6.4899822882306105e-06, | |
| "loss": 0.0545, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 6.95, | |
| "eval_accuracy": 0.9200073219842577, | |
| "eval_loss": 0.508368730545044, | |
| "eval_runtime": 15.453, | |
| "eval_samples_per_second": 353.524, | |
| "eval_steps_per_second": 44.199, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 6.327488991079118e-06, | |
| "loss": 0.0465, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "eval_accuracy": 0.9192751235584844, | |
| "eval_loss": 0.5422417521476746, | |
| "eval_runtime": 15.5091, | |
| "eval_samples_per_second": 352.244, | |
| "eval_steps_per_second": 44.039, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 6.164995693927625e-06, | |
| "loss": 0.0424, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "eval_accuracy": 0.920190371590701, | |
| "eval_loss": 0.5029627680778503, | |
| "eval_runtime": 15.5152, | |
| "eval_samples_per_second": 352.107, | |
| "eval_steps_per_second": 44.021, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 7.18, | |
| "learning_rate": 6.002502396776133e-06, | |
| "loss": 0.0317, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 7.18, | |
| "eval_accuracy": 0.9212886692293611, | |
| "eval_loss": 0.5393094420433044, | |
| "eval_runtime": 15.5001, | |
| "eval_samples_per_second": 352.449, | |
| "eval_steps_per_second": 44.064, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "learning_rate": 5.84000909962464e-06, | |
| "loss": 0.029, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "eval_accuracy": 0.9174446274940509, | |
| "eval_loss": 0.5618457794189453, | |
| "eval_runtime": 15.5012, | |
| "eval_samples_per_second": 352.424, | |
| "eval_steps_per_second": 44.061, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 7.33, | |
| "learning_rate": 5.677515802473148e-06, | |
| "loss": 0.0439, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 7.33, | |
| "eval_accuracy": 0.9194581731649277, | |
| "eval_loss": 0.5000433325767517, | |
| "eval_runtime": 15.5179, | |
| "eval_samples_per_second": 352.046, | |
| "eval_steps_per_second": 44.014, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 7.41, | |
| "learning_rate": 5.515022505321655e-06, | |
| "loss": 0.0347, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 7.41, | |
| "eval_accuracy": 0.9200073219842577, | |
| "eval_loss": 0.5093263387680054, | |
| "eval_runtime": 15.4926, | |
| "eval_samples_per_second": 352.62, | |
| "eval_steps_per_second": 44.086, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "learning_rate": 5.352529208170163e-06, | |
| "loss": 0.0425, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "eval_accuracy": 0.9174446274940509, | |
| "eval_loss": 0.5311388373374939, | |
| "eval_runtime": 15.4852, | |
| "eval_samples_per_second": 352.789, | |
| "eval_steps_per_second": 44.107, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 7.56, | |
| "learning_rate": 5.19003591101867e-06, | |
| "loss": 0.0384, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 7.56, | |
| "eval_accuracy": 0.9198242723778144, | |
| "eval_loss": 0.501017689704895, | |
| "eval_runtime": 15.5002, | |
| "eval_samples_per_second": 352.448, | |
| "eval_steps_per_second": 44.064, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 7.64, | |
| "learning_rate": 5.027542613867178e-06, | |
| "loss": 0.039, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 7.64, | |
| "eval_accuracy": 0.9209225700164745, | |
| "eval_loss": 0.5181994438171387, | |
| "eval_runtime": 15.4866, | |
| "eval_samples_per_second": 352.755, | |
| "eval_steps_per_second": 44.103, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 7.71, | |
| "learning_rate": 4.865049316715686e-06, | |
| "loss": 0.04, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 7.71, | |
| "eval_accuracy": 0.9214717188358045, | |
| "eval_loss": 0.5238317847251892, | |
| "eval_runtime": 15.4962, | |
| "eval_samples_per_second": 352.538, | |
| "eval_steps_per_second": 44.075, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "learning_rate": 4.702556019564194e-06, | |
| "loss": 0.0374, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "eval_accuracy": 0.9218378180486912, | |
| "eval_loss": 0.5560519695281982, | |
| "eval_runtime": 15.4921, | |
| "eval_samples_per_second": 352.63, | |
| "eval_steps_per_second": 44.087, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 7.87, | |
| "learning_rate": 4.540062722412701e-06, | |
| "loss": 0.0366, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 7.87, | |
| "eval_accuracy": 0.9200073219842577, | |
| "eval_loss": 0.5412092208862305, | |
| "eval_runtime": 15.4629, | |
| "eval_samples_per_second": 353.297, | |
| "eval_steps_per_second": 44.17, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 7.94, | |
| "learning_rate": 4.377569425261209e-06, | |
| "loss": 0.036, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 7.94, | |
| "eval_accuracy": 0.9212886692293611, | |
| "eval_loss": 0.5213342905044556, | |
| "eval_runtime": 15.4852, | |
| "eval_samples_per_second": 352.789, | |
| "eval_steps_per_second": 44.107, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 4.215076128109716e-06, | |
| "loss": 0.0348, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "eval_accuracy": 0.9216547684422478, | |
| "eval_loss": 0.51399165391922, | |
| "eval_runtime": 15.5008, | |
| "eval_samples_per_second": 352.434, | |
| "eval_steps_per_second": 44.062, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "learning_rate": 4.0525828309582235e-06, | |
| "loss": 0.0186, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "eval_accuracy": 0.9240344133260113, | |
| "eval_loss": 0.5692733526229858, | |
| "eval_runtime": 15.5019, | |
| "eval_samples_per_second": 352.408, | |
| "eval_steps_per_second": 44.059, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 8.17, | |
| "learning_rate": 3.8900895338067306e-06, | |
| "loss": 0.0275, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 8.17, | |
| "eval_accuracy": 0.923851363719568, | |
| "eval_loss": 0.5007224082946777, | |
| "eval_runtime": 15.4606, | |
| "eval_samples_per_second": 353.349, | |
| "eval_steps_per_second": 44.177, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 8.25, | |
| "learning_rate": 3.727596236655238e-06, | |
| "loss": 0.0219, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 8.25, | |
| "eval_accuracy": 0.9240344133260113, | |
| "eval_loss": 0.5399835705757141, | |
| "eval_runtime": 15.5486, | |
| "eval_samples_per_second": 351.35, | |
| "eval_steps_per_second": 43.927, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 8.32, | |
| "learning_rate": 3.5651029395037455e-06, | |
| "loss": 0.0238, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 8.32, | |
| "eval_accuracy": 0.9227530660809079, | |
| "eval_loss": 0.5536710023880005, | |
| "eval_runtime": 15.4711, | |
| "eval_samples_per_second": 353.109, | |
| "eval_steps_per_second": 44.147, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "learning_rate": 3.402609642352253e-06, | |
| "loss": 0.0201, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "eval_accuracy": 0.9214717188358045, | |
| "eval_loss": 0.5851342678070068, | |
| "eval_runtime": 15.5145, | |
| "eval_samples_per_second": 352.122, | |
| "eval_steps_per_second": 44.023, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "learning_rate": 3.2401163452007604e-06, | |
| "loss": 0.0253, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "eval_accuracy": 0.9216547684422478, | |
| "eval_loss": 0.5654112696647644, | |
| "eval_runtime": 15.537, | |
| "eval_samples_per_second": 351.612, | |
| "eval_steps_per_second": 43.96, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "learning_rate": 3.0776230480492682e-06, | |
| "loss": 0.0243, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "eval_accuracy": 0.9212886692293611, | |
| "eval_loss": 0.5833209156990051, | |
| "eval_runtime": 15.4958, | |
| "eval_samples_per_second": 352.547, | |
| "eval_steps_per_second": 44.076, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 8.63, | |
| "learning_rate": 2.9151297508977757e-06, | |
| "loss": 0.0298, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 8.63, | |
| "eval_accuracy": 0.9209225700164745, | |
| "eval_loss": 0.5482963919639587, | |
| "eval_runtime": 15.4637, | |
| "eval_samples_per_second": 353.28, | |
| "eval_steps_per_second": 44.168, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 8.71, | |
| "learning_rate": 2.752636453746283e-06, | |
| "loss": 0.0232, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 8.71, | |
| "eval_accuracy": 0.9214717188358045, | |
| "eval_loss": 0.5724154114723206, | |
| "eval_runtime": 15.5462, | |
| "eval_samples_per_second": 351.403, | |
| "eval_steps_per_second": 43.933, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "learning_rate": 2.5901431565947906e-06, | |
| "loss": 0.0239, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "eval_accuracy": 0.9194581731649277, | |
| "eval_loss": 0.5574042201042175, | |
| "eval_runtime": 15.4883, | |
| "eval_samples_per_second": 352.717, | |
| "eval_steps_per_second": 44.098, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 8.86, | |
| "learning_rate": 2.427649859443298e-06, | |
| "loss": 0.0263, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 8.86, | |
| "eval_accuracy": 0.9234852645066813, | |
| "eval_loss": 0.5491210222244263, | |
| "eval_runtime": 15.5424, | |
| "eval_samples_per_second": 351.489, | |
| "eval_steps_per_second": 43.944, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 8.94, | |
| "learning_rate": 2.2651565622918055e-06, | |
| "loss": 0.0333, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 8.94, | |
| "eval_accuracy": 0.9209225700164745, | |
| "eval_loss": 0.5322240591049194, | |
| "eval_runtime": 15.4522, | |
| "eval_samples_per_second": 353.541, | |
| "eval_steps_per_second": 44.201, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "learning_rate": 2.102663265140313e-06, | |
| "loss": 0.0259, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "eval_accuracy": 0.9216547684422478, | |
| "eval_loss": 0.5492663979530334, | |
| "eval_runtime": 15.4874, | |
| "eval_samples_per_second": 352.738, | |
| "eval_steps_per_second": 44.1, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 1.9401699679888204e-06, | |
| "loss": 0.0197, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "eval_accuracy": 0.9209225700164745, | |
| "eval_loss": 0.5670880675315857, | |
| "eval_runtime": 15.4679, | |
| "eval_samples_per_second": 353.183, | |
| "eval_steps_per_second": 44.156, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 9.16, | |
| "learning_rate": 1.777676670837328e-06, | |
| "loss": 0.0237, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 9.16, | |
| "eval_accuracy": 0.9209225700164745, | |
| "eval_loss": 0.5536203980445862, | |
| "eval_runtime": 15.492, | |
| "eval_samples_per_second": 352.633, | |
| "eval_steps_per_second": 44.087, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 9.24, | |
| "learning_rate": 1.6151833736858355e-06, | |
| "loss": 0.022, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 9.24, | |
| "eval_accuracy": 0.9216547684422478, | |
| "eval_loss": 0.5523190498352051, | |
| "eval_runtime": 15.499, | |
| "eval_samples_per_second": 352.475, | |
| "eval_steps_per_second": 44.067, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 9.32, | |
| "learning_rate": 1.452690076534343e-06, | |
| "loss": 0.0246, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 9.32, | |
| "eval_accuracy": 0.9220208676551346, | |
| "eval_loss": 0.5619478821754456, | |
| "eval_runtime": 15.4813, | |
| "eval_samples_per_second": 352.878, | |
| "eval_steps_per_second": 44.118, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 9.39, | |
| "learning_rate": 1.2901967793828504e-06, | |
| "loss": 0.0202, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 9.39, | |
| "eval_accuracy": 0.9227530660809079, | |
| "eval_loss": 0.5619451999664307, | |
| "eval_runtime": 15.4982, | |
| "eval_samples_per_second": 352.492, | |
| "eval_steps_per_second": 44.07, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "learning_rate": 1.127703482231358e-06, | |
| "loss": 0.0184, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "eval_accuracy": 0.9216547684422478, | |
| "eval_loss": 0.5728563070297241, | |
| "eval_runtime": 15.4841, | |
| "eval_samples_per_second": 352.814, | |
| "eval_steps_per_second": 44.11, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 9.55, | |
| "learning_rate": 9.652101850798656e-07, | |
| "loss": 0.0122, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 9.55, | |
| "eval_accuracy": 0.920190371590701, | |
| "eval_loss": 0.594571053981781, | |
| "eval_runtime": 15.5059, | |
| "eval_samples_per_second": 352.317, | |
| "eval_steps_per_second": 44.048, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 9.62, | |
| "learning_rate": 8.02716887928373e-07, | |
| "loss": 0.015, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 9.62, | |
| "eval_accuracy": 0.9214717188358045, | |
| "eval_loss": 0.601406455039978, | |
| "eval_runtime": 15.5587, | |
| "eval_samples_per_second": 351.121, | |
| "eval_steps_per_second": 43.898, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 9.7, | |
| "learning_rate": 6.402235907768806e-07, | |
| "loss": 0.0189, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 9.7, | |
| "eval_accuracy": 0.9225700164744646, | |
| "eval_loss": 0.5927833318710327, | |
| "eval_runtime": 15.5411, | |
| "eval_samples_per_second": 351.518, | |
| "eval_steps_per_second": 43.948, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 9.78, | |
| "learning_rate": 4.777302936253879e-07, | |
| "loss": 0.0194, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 9.78, | |
| "eval_accuracy": 0.9220208676551346, | |
| "eval_loss": 0.5897760987281799, | |
| "eval_runtime": 15.5041, | |
| "eval_samples_per_second": 352.359, | |
| "eval_steps_per_second": 44.053, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 9.85, | |
| "learning_rate": 3.1523699647389547e-07, | |
| "loss": 0.0219, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 9.85, | |
| "eval_accuracy": 0.9218378180486912, | |
| "eval_loss": 0.5851386189460754, | |
| "eval_runtime": 15.4835, | |
| "eval_samples_per_second": 352.826, | |
| "eval_steps_per_second": 44.111, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 9.93, | |
| "learning_rate": 1.5274369932240295e-07, | |
| "loss": 0.017, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 9.93, | |
| "eval_accuracy": 0.9218378180486912, | |
| "eval_loss": 0.5890805125236511, | |
| "eval_runtime": 15.4838, | |
| "eval_samples_per_second": 352.819, | |
| "eval_steps_per_second": 44.111, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 65470, | |
| "total_flos": 6.88976031789312e+16, | |
| "train_loss": 0.1338691882801653, | |
| "train_runtime": 14647.724, | |
| "train_samples_per_second": 71.508, | |
| "train_steps_per_second": 4.47 | |
| } | |
| ], | |
| "max_steps": 65470, | |
| "num_train_epochs": 10, | |
| "total_flos": 6.88976031789312e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |