| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9871244635193133, | |
| "eval_steps": 500, | |
| "global_step": 522, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02861230329041488, | |
| "grad_norm": 2827.3015022702225, | |
| "learning_rate": 9.766490138119515e-06, | |
| "loss": 6.1313, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.05722460658082976, | |
| "grad_norm": 516.297994115725, | |
| "learning_rate": 1.3972688495262568e-05, | |
| "loss": 3.18, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08583690987124463, | |
| "grad_norm": 10.64334537255521, | |
| "learning_rate": 1.6433156804786183e-05, | |
| "loss": 0.6962, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.11444921316165951, | |
| "grad_norm": 21.79989086330797, | |
| "learning_rate": 1.8178886852405614e-05, | |
| "loss": 0.4631, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1430615164520744, | |
| "grad_norm": 8.407923664935195, | |
| "learning_rate": 1.953298027623903e-05, | |
| "loss": 0.3384, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.17167381974248927, | |
| "grad_norm": 4.980763069359495, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2112, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.20028612303290416, | |
| "grad_norm": 3.3050323983887084, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1508, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.22889842632331903, | |
| "grad_norm": 2.747487969978562, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1331, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2575107296137339, | |
| "grad_norm": 3.8930367082100776, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1279, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2861230329041488, | |
| "grad_norm": 2.262054738771732, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1245, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.3147353361945637, | |
| "grad_norm": 2.1250650940368487, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1133, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.34334763948497854, | |
| "grad_norm": 1.770036809289755, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0945, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.3719599427753934, | |
| "grad_norm": 1.7938876086769961, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0866, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.4005722460658083, | |
| "grad_norm": 1.4764321868179242, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0827, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.4291845493562232, | |
| "grad_norm": 1.9128984486999376, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0798, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.45779685264663805, | |
| "grad_norm": 1.46483598523261, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0804, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.4864091559370529, | |
| "grad_norm": 1.6523142945768603, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0758, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.5150214592274678, | |
| "grad_norm": 1.5333568395438437, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0709, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5436337625178826, | |
| "grad_norm": 1.5031573664657636, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0687, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.5722460658082976, | |
| "grad_norm": 1.4132321325139292, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0574, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.6008583690987125, | |
| "grad_norm": 1.3745064149711035, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0572, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.6294706723891274, | |
| "grad_norm": 1.3470725333478024, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0611, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.6580829756795422, | |
| "grad_norm": 1.1315562008981583, | |
| "learning_rate": 2e-05, | |
| "loss": 0.05, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.6866952789699571, | |
| "grad_norm": 1.2282177741629088, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0525, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.7153075822603719, | |
| "grad_norm": 1.3933198446492454, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0519, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.7439198855507868, | |
| "grad_norm": 1.7829406158054193, | |
| "learning_rate": 2e-05, | |
| "loss": 0.051, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.7725321888412017, | |
| "grad_norm": 1.1948798936363785, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0514, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.8011444921316166, | |
| "grad_norm": 1.0816375150345683, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0512, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.8297567954220315, | |
| "grad_norm": 1.469354846951377, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0465, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.8583690987124464, | |
| "grad_norm": 1.2522970466753844, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0493, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.8869814020028612, | |
| "grad_norm": 1.162286189716735, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0474, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.9155937052932761, | |
| "grad_norm": 1.0718851830181713, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0415, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.944206008583691, | |
| "grad_norm": 1.0733174430217316, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0528, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.9728183118741058, | |
| "grad_norm": 1.0644789712973826, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0456, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.0014306151645207, | |
| "grad_norm": 1.1496891631410193, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0432, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.0300429184549356, | |
| "grad_norm": 1.0200369998966563, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0241, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.0586552217453504, | |
| "grad_norm": 1.3931472581826994, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0225, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 1.0872675250357653, | |
| "grad_norm": 1.0114502707229882, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0239, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.1158798283261802, | |
| "grad_norm": 0.7510204564859241, | |
| "learning_rate": 2e-05, | |
| "loss": 0.027, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 1.144492131616595, | |
| "grad_norm": 0.7295063245482636, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0235, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.17310443490701, | |
| "grad_norm": 0.856671906186716, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0247, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 1.201716738197425, | |
| "grad_norm": 0.7255694844652782, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0258, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.2303290414878398, | |
| "grad_norm": 0.9184490713297652, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0276, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 1.2589413447782547, | |
| "grad_norm": 0.818205379161712, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0219, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.2875536480686696, | |
| "grad_norm": 0.6654369429209799, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0261, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.3161659513590844, | |
| "grad_norm": 0.5829250860946364, | |
| "learning_rate": 2e-05, | |
| "loss": 0.017, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.3447782546494993, | |
| "grad_norm": 0.6661015288674467, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0228, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 1.3733905579399142, | |
| "grad_norm": 0.7177595155125043, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0178, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.402002861230329, | |
| "grad_norm": 0.8848450057764279, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0147, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 1.4306151645207439, | |
| "grad_norm": 0.6571566830429797, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0216, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.4592274678111588, | |
| "grad_norm": 0.812436234834659, | |
| "learning_rate": 2e-05, | |
| "loss": 0.023, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.4878397711015736, | |
| "grad_norm": 0.7840598860015469, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0209, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.5164520743919887, | |
| "grad_norm": 0.7844249253805873, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0205, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.5450643776824036, | |
| "grad_norm": 0.7640044613122257, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0238, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.5736766809728184, | |
| "grad_norm": 1.0261564863265702, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0271, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.6022889842632333, | |
| "grad_norm": 0.6603554019675723, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0224, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.6309012875536482, | |
| "grad_norm": 0.6112434008888443, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0201, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 1.659513590844063, | |
| "grad_norm": 0.6941562759172227, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0209, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.688125894134478, | |
| "grad_norm": 0.920122539331784, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0224, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 1.7167381974248928, | |
| "grad_norm": 0.784663743006703, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0204, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.7453505007153076, | |
| "grad_norm": 0.5371735092585386, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0175, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 1.7739628040057225, | |
| "grad_norm": 0.4569754495971157, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0191, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.8025751072961373, | |
| "grad_norm": 0.5809346149070659, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0163, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 1.8311874105865522, | |
| "grad_norm": 0.9416876917379606, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0186, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.859799713876967, | |
| "grad_norm": 0.9128407546360238, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0177, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.888412017167382, | |
| "grad_norm": 0.6404265787090032, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0142, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.9170243204577968, | |
| "grad_norm": 0.8659777729110113, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0188, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 1.9456366237482117, | |
| "grad_norm": 0.7389926990941214, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0205, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.9742489270386265, | |
| "grad_norm": 0.5010753341589564, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0173, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 2.0028612303290414, | |
| "grad_norm": 0.44585995494505515, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0176, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.0314735336194563, | |
| "grad_norm": 0.3372647662374521, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0114, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 2.060085836909871, | |
| "grad_norm": 0.5724492338755197, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0144, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.088698140200286, | |
| "grad_norm": 0.5187760378661088, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0137, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 2.117310443490701, | |
| "grad_norm": 0.5106154674345083, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0116, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.1459227467811157, | |
| "grad_norm": 0.5354245824814097, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0122, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.1745350500715306, | |
| "grad_norm": 0.5465913593959528, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0135, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.2031473533619454, | |
| "grad_norm": 0.7553346807762494, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0139, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 2.2317596566523603, | |
| "grad_norm": 0.47986647806349203, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0125, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.260371959942775, | |
| "grad_norm": 0.6926460223156353, | |
| "learning_rate": 2e-05, | |
| "loss": 0.012, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 2.28898426323319, | |
| "grad_norm": 0.7721328467064986, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0104, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.317596566523605, | |
| "grad_norm": 0.3938351782885143, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0094, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 2.34620886981402, | |
| "grad_norm": 0.7956446454256582, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0127, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.374821173104435, | |
| "grad_norm": 0.7163390976077646, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0117, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 2.40343347639485, | |
| "grad_norm": 0.40417151839328475, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0116, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.432045779685265, | |
| "grad_norm": 0.7406033829214401, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0115, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.4606580829756797, | |
| "grad_norm": 1.0372948520488305, | |
| "learning_rate": 2e-05, | |
| "loss": 0.013, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.4892703862660945, | |
| "grad_norm": 0.48303247551117084, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0121, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 2.5178826895565094, | |
| "grad_norm": 0.822531770752665, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0091, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.5464949928469243, | |
| "grad_norm": 0.5751055353850153, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0117, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 2.575107296137339, | |
| "grad_norm": 0.8111046011909318, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0138, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.603719599427754, | |
| "grad_norm": 0.5529988693729204, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0137, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 2.632331902718169, | |
| "grad_norm": 0.6426046706622803, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0128, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.6609442060085837, | |
| "grad_norm": 0.5842059243112792, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0097, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 2.6895565092989986, | |
| "grad_norm": 0.9462035289351468, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0111, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.7181688125894135, | |
| "grad_norm": 0.47730280851213186, | |
| "learning_rate": 2e-05, | |
| "loss": 0.01, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 2.7467811158798283, | |
| "grad_norm": 0.7829145546901836, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0147, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.775393419170243, | |
| "grad_norm": 0.41532702346006606, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0106, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 2.804005722460658, | |
| "grad_norm": 0.42916154288878555, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0135, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.832618025751073, | |
| "grad_norm": 0.5117471137135019, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0095, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 2.8612303290414878, | |
| "grad_norm": 0.42607884594383777, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0105, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.8898426323319026, | |
| "grad_norm": 0.5078330866142711, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0111, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 2.9184549356223175, | |
| "grad_norm": 0.872560256021644, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0096, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.9470672389127324, | |
| "grad_norm": 0.5473131169980906, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0119, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 2.9756795422031472, | |
| "grad_norm": 0.33829761962434335, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0094, | |
| "step": 520 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 522, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 10457554665472.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |