| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "global_step": 570, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.4285714285714286e-06, | |
| "loss": 1.1535, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 7.142857142857143e-06, | |
| "loss": 1.0482, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 0.9675, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 2.1428571428571428e-05, | |
| "loss": 0.8126, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 0.7321, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.7198, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 4.2857142857142856e-05, | |
| "loss": 0.785, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7282, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 4.998922515567496e-05, | |
| "loss": 0.7238, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 4.995690991048146e-05, | |
| "loss": 0.6667, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 4.99030821197584e-05, | |
| "loss": 0.6404, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 4.982778818239101e-05, | |
| "loss": 0.4274, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_f1": 0.7878385848535102, | |
| "eval_loss": 0.4914661943912506, | |
| "eval_runtime": 0.6719, | |
| "eval_samples_per_second": 150.315, | |
| "eval_steps_per_second": 19.348, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 4.97310930008156e-05, | |
| "loss": 0.5671, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 4.961307992507443e-05, | |
| "loss": 0.3926, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 4.947385068096907e-05, | |
| "loss": 0.2036, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 4.9313525282373974e-05, | |
| "loss": 0.3489, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 4.9132241927786035e-05, | |
| "loss": 0.3307, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 4.893015688119921e-05, | |
| "loss": 0.2562, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 4.870744433740688e-05, | |
| "loss": 0.322, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 4.8464296271848155e-05, | |
| "loss": 0.2665, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 4.8200922275127355e-05, | |
| "loss": 0.249, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 4.7917549372349616e-05, | |
| "loss": 0.2196, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 4.761442182742799e-05, | |
| "loss": 0.2177, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_f1": 0.848931623931624, | |
| "eval_loss": 0.4218161702156067, | |
| "eval_runtime": 0.4386, | |
| "eval_samples_per_second": 230.292, | |
| "eval_steps_per_second": 29.641, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 4.7291800932531064e-05, | |
| "loss": 0.2602, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 4.694996478285231e-05, | |
| "loss": 0.5002, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 4.6589208036895535e-05, | |
| "loss": 0.2037, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 4.620984166248288e-05, | |
| "loss": 0.1931, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 4.581219266870446e-05, | |
| "loss": 0.1799, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 4.53966038240406e-05, | |
| "loss": 0.1665, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 4.496343336089965e-05, | |
| "loss": 0.1888, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 4.4513054666826146e-05, | |
| "loss": 0.0738, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 4.4045855962645363e-05, | |
| "loss": 0.1971, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 4.3562239967821805e-05, | |
| "loss": 0.2111, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 4.306262355332006e-05, | |
| "loss": 0.0525, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 4.254743738226721e-05, | |
| "loss": 0.1154, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_f1": 0.959920634920635, | |
| "eval_loss": 0.17385226488113403, | |
| "eval_runtime": 1.7225, | |
| "eval_samples_per_second": 58.637, | |
| "eval_steps_per_second": 7.547, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 4.201712553872658e-05, | |
| "loss": 0.2055, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 4.147214514490278e-05, | |
| "loss": 0.0767, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 4.0912965967108125e-05, | |
| "loss": 0.0224, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 4.034007001082985e-05, | |
| "loss": 0.1034, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 3.975395110524742e-05, | |
| "loss": 0.0192, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 3.9155114477557933e-05, | |
| "loss": 0.069, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 3.854407631747654e-05, | |
| "loss": 0.0203, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 3.792136333228735e-05, | |
| "loss": 0.2089, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.77, | |
| "learning_rate": 3.728751229282836e-05, | |
| "loss": 0.1181, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 3.664306957080159e-05, | |
| "loss": 0.0077, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 3.598859066780754e-05, | |
| "loss": 0.1192, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_f1": 0.9592413236481033, | |
| "eval_loss": 0.15608876943588257, | |
| "eval_runtime": 0.4317, | |
| "eval_samples_per_second": 233.965, | |
| "eval_steps_per_second": 30.114, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 3.5324639736509714e-05, | |
| "loss": 0.0058, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 3.4651789094342044e-05, | |
| "loss": 0.038, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 3.39706187301784e-05, | |
| "loss": 0.0024, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "learning_rate": 3.3281715804389403e-05, | |
| "loss": 0.0277, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "learning_rate": 3.258567414271748e-05, | |
| "loss": 0.0023, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 3.18830937244065e-05, | |
| "loss": 0.145, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "learning_rate": 3.117458016502711e-05, | |
| "loss": 0.0818, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 3.046074419444366e-05, | |
| "loss": 0.0089, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 2.9742201130372693e-05, | |
| "loss": 0.0026, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "learning_rate": 2.901957034798671e-05, | |
| "loss": 0.0032, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "learning_rate": 2.8293474746020472e-05, | |
| "loss": 0.0339, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 2.756454020984009e-05, | |
| "loss": 0.0022, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_f1": 0.9592413236481033, | |
| "eval_loss": 0.09981890022754669, | |
| "eval_runtime": 0.4318, | |
| "eval_samples_per_second": 233.88, | |
| "eval_steps_per_second": 30.103, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "learning_rate": 2.68333950719376e-05, | |
| "loss": 0.0023, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 5.18, | |
| "learning_rate": 2.6100669570316195e-05, | |
| "loss": 0.0022, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "learning_rate": 2.5366995305232916e-05, | |
| "loss": 0.0016, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "learning_rate": 2.463300469476709e-05, | |
| "loss": 0.0012, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "learning_rate": 2.3899330429683807e-05, | |
| "loss": 0.0011, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "learning_rate": 2.3166604928062406e-05, | |
| "loss": 0.002, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "learning_rate": 2.243545979015992e-05, | |
| "loss": 0.0051, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "learning_rate": 2.1706525253979534e-05, | |
| "loss": 0.0169, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "learning_rate": 2.0980429652013297e-05, | |
| "loss": 0.0925, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "learning_rate": 2.025779886962731e-05, | |
| "loss": 0.0008, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "learning_rate": 1.9539255805556344e-05, | |
| "loss": 0.0009, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_f1": 0.9797514033680834, | |
| "eval_loss": 0.10345583409070969, | |
| "eval_runtime": 0.411, | |
| "eval_samples_per_second": 245.716, | |
| "eval_steps_per_second": 31.627, | |
| "step": 342 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 1.8825419834972902e-05, | |
| "loss": 0.0508, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 6.14, | |
| "learning_rate": 1.811690627559351e-05, | |
| "loss": 0.0014, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 6.23, | |
| "learning_rate": 1.7414325857282526e-05, | |
| "loss": 0.0014, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 6.32, | |
| "learning_rate": 1.6718284195610606e-05, | |
| "loss": 0.0014, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 1.6029381269821604e-05, | |
| "loss": 0.0011, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 6.49, | |
| "learning_rate": 1.534821090565796e-05, | |
| "loss": 0.0008, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 6.58, | |
| "learning_rate": 1.4675360263490295e-05, | |
| "loss": 0.0009, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 1.4011409332192472e-05, | |
| "loss": 0.0043, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "learning_rate": 1.335693042919841e-05, | |
| "loss": 0.0006, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 6.84, | |
| "learning_rate": 1.2712487707171645e-05, | |
| "loss": 0.0246, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 6.93, | |
| "learning_rate": 1.2078636667712649e-05, | |
| "loss": 0.0006, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_f1": 0.9797514033680834, | |
| "eval_loss": 0.12670956552028656, | |
| "eval_runtime": 0.4031, | |
| "eval_samples_per_second": 250.582, | |
| "eval_steps_per_second": 32.253, | |
| "step": 399 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 1.1455923682523475e-05, | |
| "loss": 0.0007, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 7.11, | |
| "learning_rate": 1.0844885522442074e-05, | |
| "loss": 0.0007, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 7.19, | |
| "learning_rate": 1.0246048894752589e-05, | |
| "loss": 0.0006, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 7.28, | |
| "learning_rate": 9.659929989170154e-06, | |
| "loss": 0.0006, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 9.087034032891883e-06, | |
| "loss": 0.0007, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 7.46, | |
| "learning_rate": 8.527854855097225e-06, | |
| "loss": 0.0005, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 7.54, | |
| "learning_rate": 7.982874461273438e-06, | |
| "loss": 0.0005, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 7.63, | |
| "learning_rate": 7.452562617732794e-06, | |
| "loss": 0.0625, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "learning_rate": 6.93737644667995e-06, | |
| "loss": 0.0005, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 7.81, | |
| "learning_rate": 6.4377600321782e-06, | |
| "loss": 0.0005, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 7.89, | |
| "learning_rate": 5.954144037354645e-06, | |
| "loss": 0.0006, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 7.98, | |
| "learning_rate": 5.486945333173851e-06, | |
| "loss": 0.0005, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_f1": 0.9797514033680834, | |
| "eval_loss": 0.1428430825471878, | |
| "eval_runtime": 0.4454, | |
| "eval_samples_per_second": 226.778, | |
| "eval_steps_per_second": 29.189, | |
| "step": 456 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 5.036566639100351e-06, | |
| "loss": 0.0096, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 8.16, | |
| "learning_rate": 4.603396175959404e-06, | |
| "loss": 0.0005, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 8.25, | |
| "learning_rate": 4.187807331295549e-06, | |
| "loss": 0.0005, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 3.7901583375171273e-06, | |
| "loss": 0.0005, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 8.42, | |
| "learning_rate": 3.4107919631044732e-06, | |
| "loss": 0.0004, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 8.51, | |
| "learning_rate": 3.0500352171476897e-06, | |
| "loss": 0.0005, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 8.6, | |
| "learning_rate": 2.708199067468939e-06, | |
| "loss": 0.0005, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 8.68, | |
| "learning_rate": 2.385578172572009e-06, | |
| "loss": 0.0004, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "learning_rate": 2.0824506276503897e-06, | |
| "loss": 0.0005, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 8.86, | |
| "learning_rate": 1.7990777248726442e-06, | |
| "loss": 0.0005, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 8.95, | |
| "learning_rate": 1.5357037281518522e-06, | |
| "loss": 0.0005, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_f1": 0.9797514033680834, | |
| "eval_loss": 0.13912993669509888, | |
| "eval_runtime": 0.43, | |
| "eval_samples_per_second": 234.9, | |
| "eval_steps_per_second": 30.235, | |
| "step": 513 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 1.2925556625931173e-06, | |
| "loss": 0.0004, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 9.12, | |
| "learning_rate": 1.0698431188007952e-06, | |
| "loss": 0.0005, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 9.21, | |
| "learning_rate": 8.677580722139672e-07, | |
| "loss": 0.0004, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "learning_rate": 6.864747176260289e-07, | |
| "loss": 0.0004, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 9.39, | |
| "learning_rate": 5.261493190309303e-07, | |
| "loss": 0.0005, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 9.47, | |
| "learning_rate": 3.8692007492557024e-07, | |
| "loss": 0.0004, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 9.56, | |
| "learning_rate": 2.6890699918440676e-07, | |
| "loss": 0.0005, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 9.65, | |
| "learning_rate": 1.7221181760899152e-07, | |
| "loss": 0.0004, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 9.74, | |
| "learning_rate": 9.691788024160376e-08, | |
| "loss": 0.0004, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 9.82, | |
| "learning_rate": 4.3090089518540987e-08, | |
| "loss": 0.0004, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 9.91, | |
| "learning_rate": 1.0774844325039946e-08, | |
| "loss": 0.0004, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.0005, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_f1": 0.9797514033680834, | |
| "eval_loss": 0.1402171403169632, | |
| "eval_runtime": 1.7273, | |
| "eval_samples_per_second": 58.472, | |
| "eval_steps_per_second": 7.526, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 570, | |
| "total_flos": 200268366788952.0, | |
| "train_loss": 0.13698372087817182, | |
| "train_runtime": 202.1369, | |
| "train_samples_per_second": 44.92, | |
| "train_steps_per_second": 2.82 | |
| } | |
| ], | |
| "max_steps": 570, | |
| "num_train_epochs": 10, | |
| "total_flos": 200268366788952.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |