| { |
| "best_metric": 1.061107873916626, |
| "best_model_checkpoint": "saves/LLaMA2-13B/lora/train_2024-01-03-02-43-22/checkpoint-400", |
| "epoch": 2.9966850828729283, |
| "eval_steps": 100, |
| "global_step": 678, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.02, |
| "learning_rate": 0.00019997316318671806, |
| "loss": 1.4828, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.04, |
| "learning_rate": 0.00019989266715116316, |
| "loss": 1.1204, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.07, |
| "learning_rate": 0.00019975855509847686, |
| "loss": 1.1449, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.09, |
| "learning_rate": 0.00019957089901146148, |
| "loss": 1.1126, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.11, |
| "learning_rate": 0.00019932979961194435, |
| "loss": 1.1048, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.13, |
| "learning_rate": 0.0001990353863067169, |
| "loss": 1.2161, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.15, |
| "learning_rate": 0.00019868781711807705, |
| "loss": 1.0051, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.18, |
| "learning_rate": 0.00019828727859901317, |
| "loss": 1.0314, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.2, |
| "learning_rate": 0.00019783398573307428, |
| "loss": 1.0718, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.22, |
| "learning_rate": 0.00019732818181898045, |
| "loss": 1.1571, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.24, |
| "learning_rate": 0.0001967701383400357, |
| "loss": 1.0602, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.27, |
| "learning_rate": 0.0001961601548184129, |
| "loss": 1.1215, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.29, |
| "learning_rate": 0.00019549855865438965, |
| "loss": 1.1434, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.31, |
| "learning_rate": 0.00019478570495062037, |
| "loss": 1.0844, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.33, |
| "learning_rate": 0.00019402197632153992, |
| "loss": 1.1123, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.35, |
| "learning_rate": 0.00019320778268800066, |
| "loss": 1.1054, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.38, |
| "learning_rate": 0.00019234356105725297, |
| "loss": 1.0368, |
| "step": 85 |
| }, |
| { |
| "epoch": 0.4, |
| "learning_rate": 0.00019142977528838762, |
| "loss": 1.1195, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.42, |
| "learning_rate": 0.00019046691584336577, |
| "loss": 1.0778, |
| "step": 95 |
| }, |
| { |
| "epoch": 0.44, |
| "learning_rate": 0.0001894554995237703, |
| "loss": 1.0228, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.44, |
| "eval_loss": 1.0887844562530518, |
| "eval_runtime": 75.527, |
| "eval_samples_per_second": 11.982, |
| "eval_steps_per_second": 1.509, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.46, |
| "learning_rate": 0.0001883960691934196, |
| "loss": 1.1156, |
| "step": 105 |
| }, |
| { |
| "epoch": 0.49, |
| "learning_rate": 0.00018728919348699283, |
| "loss": 1.0259, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.51, |
| "learning_rate": 0.00018613546650482322, |
| "loss": 1.0112, |
| "step": 115 |
| }, |
| { |
| "epoch": 0.53, |
| "learning_rate": 0.00018493550749402278, |
| "loss": 1.0279, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.55, |
| "learning_rate": 0.00018368996051610986, |
| "loss": 1.105, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.57, |
| "learning_rate": 0.00018239949410131802, |
| "loss": 1.041, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.6, |
| "learning_rate": 0.00018106480088977172, |
| "loss": 1.0475, |
| "step": 135 |
| }, |
| { |
| "epoch": 0.62, |
| "learning_rate": 0.00017968659725972112, |
| "loss": 1.0241, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.64, |
| "learning_rate": 0.00017826562294303585, |
| "loss": 1.0071, |
| "step": 145 |
| }, |
| { |
| "epoch": 0.66, |
| "learning_rate": 0.0001768026406281642, |
| "loss": 0.9714, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.69, |
| "learning_rate": 0.00017529843555077066, |
| "loss": 1.0376, |
| "step": 155 |
| }, |
| { |
| "epoch": 0.71, |
| "learning_rate": 0.00017375381507227108, |
| "loss": 0.9806, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.73, |
| "learning_rate": 0.00017216960824649303, |
| "loss": 1.1022, |
| "step": 165 |
| }, |
| { |
| "epoch": 0.75, |
| "learning_rate": 0.00017054666537469213, |
| "loss": 0.9723, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.77, |
| "learning_rate": 0.00016888585754916476, |
| "loss": 1.1081, |
| "step": 175 |
| }, |
| { |
| "epoch": 0.8, |
| "learning_rate": 0.00016718807618570106, |
| "loss": 1.0525, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.82, |
| "learning_rate": 0.00016545423254513004, |
| "loss": 1.0761, |
| "step": 185 |
| }, |
| { |
| "epoch": 0.84, |
| "learning_rate": 0.00016368525724421248, |
| "loss": 1.1493, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.86, |
| "learning_rate": 0.00016188209975614542, |
| "loss": 0.9942, |
| "step": 195 |
| }, |
| { |
| "epoch": 0.88, |
| "learning_rate": 0.00016004572790094535, |
| "loss": 1.0068, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.88, |
| "eval_loss": 1.0737706422805786, |
| "eval_runtime": 75.4895, |
| "eval_samples_per_second": 11.988, |
| "eval_steps_per_second": 1.51, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.91, |
| "learning_rate": 0.00015817712732598413, |
| "loss": 1.0579, |
| "step": 205 |
| }, |
| { |
| "epoch": 0.93, |
| "learning_rate": 0.00015627730097695638, |
| "loss": 1.0985, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.95, |
| "learning_rate": 0.00015434726855956206, |
| "loss": 1.071, |
| "step": 215 |
| }, |
| { |
| "epoch": 0.97, |
| "learning_rate": 0.00015238806599219336, |
| "loss": 1.0617, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.99, |
| "learning_rate": 0.00015040074484992, |
| "loss": 0.9789, |
| "step": 225 |
| }, |
| { |
| "epoch": 1.02, |
| "learning_rate": 0.00014838637180007047, |
| "loss": 1.072, |
| "step": 230 |
| }, |
| { |
| "epoch": 1.04, |
| "learning_rate": 0.00014634602802971312, |
| "loss": 1.0098, |
| "step": 235 |
| }, |
| { |
| "epoch": 1.06, |
| "learning_rate": 0.00014428080866534396, |
| "loss": 1.0222, |
| "step": 240 |
| }, |
| { |
| "epoch": 1.08, |
| "learning_rate": 0.0001421918221850923, |
| "loss": 0.9734, |
| "step": 245 |
| }, |
| { |
| "epoch": 1.1, |
| "learning_rate": 0.00014008018982376044, |
| "loss": 0.9364, |
| "step": 250 |
| }, |
| { |
| "epoch": 1.13, |
| "learning_rate": 0.00013794704497101655, |
| "loss": 0.9735, |
| "step": 255 |
| }, |
| { |
| "epoch": 1.15, |
| "learning_rate": 0.00013579353256306287, |
| "loss": 1.0933, |
| "step": 260 |
| }, |
| { |
| "epoch": 1.17, |
| "learning_rate": 0.00013362080846810725, |
| "loss": 0.9764, |
| "step": 265 |
| }, |
| { |
| "epoch": 1.19, |
| "learning_rate": 0.00013143003886596669, |
| "loss": 1.0584, |
| "step": 270 |
| }, |
| { |
| "epoch": 1.22, |
| "learning_rate": 0.00012922239962213637, |
| "loss": 0.9116, |
| "step": 275 |
| }, |
| { |
| "epoch": 1.24, |
| "learning_rate": 0.00012699907565665982, |
| "loss": 1.0587, |
| "step": 280 |
| }, |
| { |
| "epoch": 1.26, |
| "learning_rate": 0.00012476126030813963, |
| "loss": 0.9342, |
| "step": 285 |
| }, |
| { |
| "epoch": 1.28, |
| "learning_rate": 0.00012251015469322916, |
| "loss": 0.9826, |
| "step": 290 |
| }, |
| { |
| "epoch": 1.3, |
| "learning_rate": 0.00012024696706194967, |
| "loss": 1.0656, |
| "step": 295 |
| }, |
| { |
| "epoch": 1.33, |
| "learning_rate": 0.00011797291214917881, |
| "loss": 1.0218, |
| "step": 300 |
| }, |
| { |
| "epoch": 1.33, |
| "eval_loss": 1.06796395778656, |
| "eval_runtime": 75.5412, |
| "eval_samples_per_second": 11.98, |
| "eval_steps_per_second": 1.509, |
| "step": 300 |
| }, |
| { |
| "epoch": 1.35, |
| "learning_rate": 0.00011568921052265836, |
| "loss": 0.9915, |
| "step": 305 |
| }, |
| { |
| "epoch": 1.37, |
| "learning_rate": 0.00011339708792787119, |
| "loss": 0.9592, |
| "step": 310 |
| }, |
| { |
| "epoch": 1.39, |
| "learning_rate": 0.00011109777463013915, |
| "loss": 0.9875, |
| "step": 315 |
| }, |
| { |
| "epoch": 1.41, |
| "learning_rate": 0.00010879250475429523, |
| "loss": 1.1583, |
| "step": 320 |
| }, |
| { |
| "epoch": 1.44, |
| "learning_rate": 0.00010648251562228386, |
| "loss": 0.977, |
| "step": 325 |
| }, |
| { |
| "epoch": 1.46, |
| "learning_rate": 0.00010416904708904548, |
| "loss": 1.041, |
| "step": 330 |
| }, |
| { |
| "epoch": 1.48, |
| "learning_rate": 0.00010185334087704124, |
| "loss": 0.972, |
| "step": 335 |
| }, |
| { |
| "epoch": 1.5, |
| "learning_rate": 9.953663990977568e-05, |
| "loss": 1.0115, |
| "step": 340 |
| }, |
| { |
| "epoch": 1.52, |
| "learning_rate": 9.722018764467461e-05, |
| "loss": 1.0078, |
| "step": 345 |
| }, |
| { |
| "epoch": 1.55, |
| "learning_rate": 9.490522740567633e-05, |
| "loss": 0.955, |
| "step": 350 |
| }, |
| { |
| "epoch": 1.57, |
| "learning_rate": 9.259300171589456e-05, |
| "loss": 0.9719, |
| "step": 355 |
| }, |
| { |
| "epoch": 1.59, |
| "learning_rate": 9.028475163071141e-05, |
| "loss": 0.871, |
| "step": 360 |
| }, |
| { |
| "epoch": 1.61, |
| "learning_rate": 8.798171607165778e-05, |
| "loss": 0.9568, |
| "step": 365 |
| }, |
| { |
| "epoch": 1.64, |
| "learning_rate": 8.568513116143919e-05, |
| "loss": 0.9466, |
| "step": 370 |
| }, |
| { |
| "epoch": 1.66, |
| "learning_rate": 8.339622956046417e-05, |
| "loss": 0.8967, |
| "step": 375 |
| }, |
| { |
| "epoch": 1.68, |
| "learning_rate": 8.111623980523035e-05, |
| "loss": 0.9465, |
| "step": 380 |
| }, |
| { |
| "epoch": 1.7, |
| "learning_rate": 7.884638564892472e-05, |
| "loss": 0.9499, |
| "step": 385 |
| }, |
| { |
| "epoch": 1.72, |
| "learning_rate": 7.658788540459062e-05, |
| "loss": 0.8955, |
| "step": 390 |
| }, |
| { |
| "epoch": 1.75, |
| "learning_rate": 7.434195129121518e-05, |
| "loss": 0.8982, |
| "step": 395 |
| }, |
| { |
| "epoch": 1.77, |
| "learning_rate": 7.210978878308729e-05, |
| "loss": 1.084, |
| "step": 400 |
| }, |
| { |
| "epoch": 1.77, |
| "eval_loss": 1.061107873916626, |
| "eval_runtime": 75.5635, |
| "eval_samples_per_second": 11.977, |
| "eval_steps_per_second": 1.509, |
| "step": 400 |
| }, |
| { |
| "epoch": 1.79, |
| "learning_rate": 6.989259596277582e-05, |
| "loss": 0.8936, |
| "step": 405 |
| }, |
| { |
| "epoch": 1.81, |
| "learning_rate": 6.76915628780754e-05, |
| "loss": 0.8794, |
| "step": 410 |
| }, |
| { |
| "epoch": 1.83, |
| "learning_rate": 6.55078709032644e-05, |
| "loss": 0.9631, |
| "step": 415 |
| }, |
| { |
| "epoch": 1.86, |
| "learning_rate": 6.334269210501875e-05, |
| "loss": 0.9636, |
| "step": 420 |
| }, |
| { |
| "epoch": 1.88, |
| "learning_rate": 6.119718861332098e-05, |
| "loss": 0.9761, |
| "step": 425 |
| }, |
| { |
| "epoch": 1.9, |
| "learning_rate": 5.9072511997703226e-05, |
| "loss": 0.9732, |
| "step": 430 |
| }, |
| { |
| "epoch": 1.92, |
| "learning_rate": 5.696980264915777e-05, |
| "loss": 0.995, |
| "step": 435 |
| }, |
| { |
| "epoch": 1.94, |
| "learning_rate": 5.489018916804813e-05, |
| "loss": 1.0037, |
| "step": 440 |
| }, |
| { |
| "epoch": 1.97, |
| "learning_rate": 5.283478775834811e-05, |
| "loss": 0.8503, |
| "step": 445 |
| }, |
| { |
| "epoch": 1.99, |
| "learning_rate": 5.080470162853472e-05, |
| "loss": 0.8914, |
| "step": 450 |
| }, |
| { |
| "epoch": 2.01, |
| "learning_rate": 4.880102039945624e-05, |
| "loss": 0.9531, |
| "step": 455 |
| }, |
| { |
| "epoch": 2.03, |
| "learning_rate": 4.6824819519493057e-05, |
| "loss": 0.9294, |
| "step": 460 |
| }, |
| { |
| "epoch": 2.06, |
| "learning_rate": 4.487715968732568e-05, |
| "loss": 0.8584, |
| "step": 465 |
| }, |
| { |
| "epoch": 2.08, |
| "learning_rate": 4.29590862826191e-05, |
| "loss": 0.9285, |
| "step": 470 |
| }, |
| { |
| "epoch": 2.1, |
| "learning_rate": 4.107162880492984e-05, |
| "loss": 0.8742, |
| "step": 475 |
| }, |
| { |
| "epoch": 2.12, |
| "learning_rate": 3.921580032113602e-05, |
| "loss": 0.7903, |
| "step": 480 |
| }, |
| { |
| "epoch": 2.14, |
| "learning_rate": 3.739259692168764e-05, |
| "loss": 0.8287, |
| "step": 485 |
| }, |
| { |
| "epoch": 2.17, |
| "learning_rate": 3.560299718596889e-05, |
| "loss": 0.8967, |
| "step": 490 |
| }, |
| { |
| "epoch": 2.19, |
| "learning_rate": 3.3847961657058845e-05, |
| "loss": 0.841, |
| "step": 495 |
| }, |
| { |
| "epoch": 2.21, |
| "learning_rate": 3.212843232617343e-05, |
| "loss": 0.897, |
| "step": 500 |
| }, |
| { |
| "epoch": 2.21, |
| "eval_loss": 1.0756696462631226, |
| "eval_runtime": 75.5409, |
| "eval_samples_per_second": 11.98, |
| "eval_steps_per_second": 1.509, |
| "step": 500 |
| }, |
| { |
| "epoch": 2.23, |
| "learning_rate": 3.0445332127064275e-05, |
| "loss": 0.9455, |
| "step": 505 |
| }, |
| { |
| "epoch": 2.25, |
| "learning_rate": 2.879956444064703e-05, |
| "loss": 0.9292, |
| "step": 510 |
| }, |
| { |
| "epoch": 2.28, |
| "learning_rate": 2.7192012610123774e-05, |
| "loss": 0.9016, |
| "step": 515 |
| }, |
| { |
| "epoch": 2.3, |
| "learning_rate": 2.5623539466860813e-05, |
| "loss": 0.937, |
| "step": 520 |
| }, |
| { |
| "epoch": 2.32, |
| "learning_rate": 2.409498686727587e-05, |
| "loss": 0.8348, |
| "step": 525 |
| }, |
| { |
| "epoch": 2.34, |
| "learning_rate": 2.2607175240983026e-05, |
| "loss": 0.8387, |
| "step": 530 |
| }, |
| { |
| "epoch": 2.36, |
| "learning_rate": 2.1160903150438605e-05, |
| "loss": 0.7736, |
| "step": 535 |
| }, |
| { |
| "epoch": 2.39, |
| "learning_rate": 1.9756946862323535e-05, |
| "loss": 0.8672, |
| "step": 540 |
| }, |
| { |
| "epoch": 2.41, |
| "learning_rate": 1.839605993089307e-05, |
| "loss": 0.8688, |
| "step": 545 |
| }, |
| { |
| "epoch": 2.43, |
| "learning_rate": 1.707897279351671e-05, |
| "loss": 0.8631, |
| "step": 550 |
| }, |
| { |
| "epoch": 2.45, |
| "learning_rate": 1.580639237862608e-05, |
| "loss": 0.9171, |
| "step": 555 |
| }, |
| { |
| "epoch": 2.48, |
| "learning_rate": 1.4579001726280828e-05, |
| "loss": 0.9134, |
| "step": 560 |
| }, |
| { |
| "epoch": 2.5, |
| "learning_rate": 1.339745962155613e-05, |
| "loss": 0.8363, |
| "step": 565 |
| }, |
| { |
| "epoch": 2.52, |
| "learning_rate": 1.2262400240949023e-05, |
| "loss": 0.8862, |
| "step": 570 |
| }, |
| { |
| "epoch": 2.54, |
| "learning_rate": 1.1174432811992685e-05, |
| "loss": 0.9154, |
| "step": 575 |
| }, |
| { |
| "epoch": 2.56, |
| "learning_rate": 1.013414128626211e-05, |
| "loss": 0.9423, |
| "step": 580 |
| }, |
| { |
| "epoch": 2.59, |
| "learning_rate": 9.142084025945984e-06, |
| "loss": 0.8462, |
| "step": 585 |
| }, |
| { |
| "epoch": 2.61, |
| "learning_rate": 8.19879350415349e-06, |
| "loss": 0.8019, |
| "step": 590 |
| }, |
| { |
| "epoch": 2.63, |
| "learning_rate": 7.30477601911671e-06, |
| "loss": 0.861, |
| "step": 595 |
| }, |
| { |
| "epoch": 2.65, |
| "learning_rate": 6.460511422441984e-06, |
| "loss": 0.9597, |
| "step": 600 |
| }, |
| { |
| "epoch": 2.65, |
| "eval_loss": 1.0745042562484741, |
| "eval_runtime": 75.5387, |
| "eval_samples_per_second": 11.981, |
| "eval_steps_per_second": 1.509, |
| "step": 600 |
| }, |
| { |
| "epoch": 2.67, |
| "learning_rate": 5.66645286155616e-06, |
| "loss": 0.7797, |
| "step": 605 |
| }, |
| { |
| "epoch": 2.7, |
| "learning_rate": 4.923026536485875e-06, |
| "loss": 0.9222, |
| "step": 610 |
| }, |
| { |
| "epoch": 2.72, |
| "learning_rate": 4.230631471100655e-06, |
| "loss": 0.9732, |
| "step": 615 |
| }, |
| { |
| "epoch": 2.74, |
| "learning_rate": 3.5896392989422377e-06, |
| "loss": 0.7366, |
| "step": 620 |
| }, |
| { |
| "epoch": 2.76, |
| "learning_rate": 3.000394063755396e-06, |
| "loss": 0.9243, |
| "step": 625 |
| }, |
| { |
| "epoch": 2.78, |
| "learning_rate": 2.4632120348272003e-06, |
| "loss": 0.9263, |
| "step": 630 |
| }, |
| { |
| "epoch": 2.81, |
| "learning_rate": 1.9783815372338423e-06, |
| "loss": 0.8541, |
| "step": 635 |
| }, |
| { |
| "epoch": 2.83, |
| "learning_rate": 1.5461627970860814e-06, |
| "loss": 0.828, |
| "step": 640 |
| }, |
| { |
| "epoch": 2.85, |
| "learning_rate": 1.1667878018564171e-06, |
| "loss": 0.8446, |
| "step": 645 |
| }, |
| { |
| "epoch": 2.87, |
| "learning_rate": 8.404601758630892e-07, |
| "loss": 0.7749, |
| "step": 650 |
| }, |
| { |
| "epoch": 2.9, |
| "learning_rate": 5.673550709774267e-07, |
| "loss": 0.8738, |
| "step": 655 |
| }, |
| { |
| "epoch": 2.92, |
| "learning_rate": 3.4761907261356976e-07, |
| "loss": 0.895, |
| "step": 660 |
| }, |
| { |
| "epoch": 2.94, |
| "learning_rate": 1.813701210506946e-07, |
| "loss": 0.8552, |
| "step": 665 |
| }, |
| { |
| "epoch": 2.96, |
| "learning_rate": 6.869744813023937e-08, |
| "loss": 0.8475, |
| "step": 670 |
| }, |
| { |
| "epoch": 2.98, |
| "learning_rate": 9.661529361892907e-09, |
| "loss": 0.9076, |
| "step": 675 |
| }, |
| { |
| "epoch": 3.0, |
| "step": 678, |
| "total_flos": 1.5685942606774272e+17, |
| "train_loss": 0.9757168982232918, |
| "train_runtime": 3618.1154, |
| "train_samples_per_second": 3.0, |
| "train_steps_per_second": 0.187 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 678, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 100, |
| "total_flos": 1.5685942606774272e+17, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
