| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9994869163673679, | |
| "eval_steps": 500, | |
| "global_step": 974, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 0.4224665734985152, | |
| "learning_rate": 2.040816326530612e-06, | |
| "loss": 1.0622, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.5284465715139516, | |
| "learning_rate": 1.0204081632653061e-05, | |
| "loss": 0.9796, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.4800413465398026, | |
| "learning_rate": 2.0408163265306123e-05, | |
| "loss": 1.0297, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.2816340466714462, | |
| "learning_rate": 3.061224489795919e-05, | |
| "loss": 1.0176, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.28168990888908674, | |
| "learning_rate": 4.0816326530612245e-05, | |
| "loss": 0.9993, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.1553913296147897, | |
| "learning_rate": 5.102040816326531e-05, | |
| "loss": 0.9358, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.18142210930647196, | |
| "learning_rate": 6.122448979591838e-05, | |
| "loss": 0.9419, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.23481370895189654, | |
| "learning_rate": 7.142857142857143e-05, | |
| "loss": 0.8759, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.14980692548924152, | |
| "learning_rate": 8.163265306122449e-05, | |
| "loss": 0.883, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.133576680908321, | |
| "learning_rate": 9.183673469387756e-05, | |
| "loss": 0.862, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.12532996035673116, | |
| "learning_rate": 0.00010204081632653062, | |
| "loss": 0.9202, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.13932219083428563, | |
| "learning_rate": 0.00011224489795918367, | |
| "loss": 0.8846, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.12328088246202568, | |
| "learning_rate": 0.00012244897959183676, | |
| "loss": 0.8614, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.14214840640740986, | |
| "learning_rate": 0.0001326530612244898, | |
| "loss": 0.8456, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.14034095762109017, | |
| "learning_rate": 0.00014285714285714287, | |
| "loss": 0.8278, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.1562397541717682, | |
| "learning_rate": 0.0001530612244897959, | |
| "loss": 0.8268, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.13834727022919813, | |
| "learning_rate": 0.00016326530612244898, | |
| "loss": 0.8246, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.13195301400647858, | |
| "learning_rate": 0.00017346938775510205, | |
| "loss": 0.8858, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.1476813054333854, | |
| "learning_rate": 0.00018367346938775512, | |
| "loss": 0.8734, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.1326769165243531, | |
| "learning_rate": 0.00019387755102040816, | |
| "loss": 0.8431, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.11831952224955868, | |
| "learning_rate": 0.0001999974277115551, | |
| "loss": 0.8467, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.13613884130299994, | |
| "learning_rate": 0.00019996849098629418, | |
| "loss": 0.7846, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.13456966066319215, | |
| "learning_rate": 0.00019990741151022301, | |
| "loss": 0.8007, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.1679031230382061, | |
| "learning_rate": 0.0001998142089221534, | |
| "loss": 0.8864, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.1195588588744713, | |
| "learning_rate": 0.0001996889131894033, | |
| "loss": 0.8526, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.1316308906877641, | |
| "learning_rate": 0.00019953156459816179, | |
| "loss": 0.8104, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.1295896362357539, | |
| "learning_rate": 0.0001993422137405354, | |
| "loss": 0.8584, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.14146193582571115, | |
| "learning_rate": 0.00019912092149828174, | |
| "loss": 0.8828, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.15128325355835132, | |
| "learning_rate": 0.00019886775902323405, | |
| "loss": 0.8569, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.14649488912781936, | |
| "learning_rate": 0.00019858280771442385, | |
| "loss": 0.8549, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.18943998152343347, | |
| "learning_rate": 0.00019826615919190887, | |
| "loss": 0.8605, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.124667248916099, | |
| "learning_rate": 0.00019791791526731445, | |
| "loss": 0.8382, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.22445656429732966, | |
| "learning_rate": 0.00019753818791109828, | |
| "loss": 0.8163, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.14098918819116132, | |
| "learning_rate": 0.0001971270992165486, | |
| "loss": 0.778, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.13339319403763952, | |
| "learning_rate": 0.00019668478136052774, | |
| "loss": 0.8138, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.13961481566344255, | |
| "learning_rate": 0.0001962113765609735, | |
| "loss": 0.7989, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.13087704871428119, | |
| "learning_rate": 0.0001957070370311717, | |
| "loss": 0.8503, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.14182563419142846, | |
| "learning_rate": 0.00019517192493081565, | |
| "loss": 0.8253, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.12666210841941644, | |
| "learning_rate": 0.00019460621231386676, | |
| "loss": 0.8828, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.1271725385943482, | |
| "learning_rate": 0.00019401008107323455, | |
| "loss": 0.8586, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.13336647812691615, | |
| "learning_rate": 0.0001933837228822925, | |
| "loss": 0.8086, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.13715468934306543, | |
| "learning_rate": 0.0001927273391332499, | |
| "loss": 0.8562, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.13349689371469725, | |
| "learning_rate": 0.00019204114087239806, | |
| "loss": 0.8601, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.14685924776626536, | |
| "learning_rate": 0.00019132534873225323, | |
| "loss": 0.8373, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.1350372733252807, | |
| "learning_rate": 0.00019058019286061665, | |
| "loss": 0.8331, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.12803447380882063, | |
| "learning_rate": 0.00018980591284657535, | |
| "loss": 0.8365, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.1334208833129311, | |
| "learning_rate": 0.00018900275764346768, | |
| "loss": 0.8322, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.14702003545455958, | |
| "learning_rate": 0.0001881709854888372, | |
| "loss": 0.8217, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.3214308870564164, | |
| "learning_rate": 0.00018731086382140226, | |
| "loss": 0.8154, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.14661590768195695, | |
| "learning_rate": 0.00018642266919506644, | |
| "loss": 0.8471, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.15237243420118865, | |
| "learning_rate": 0.00018550668718999872, | |
| "loss": 0.795, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.1185694028090917, | |
| "learning_rate": 0.0001845632123208111, | |
| "loss": 0.8544, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.13250010062470685, | |
| "learning_rate": 0.0001835925479418637, | |
| "loss": 0.8608, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.11443256555837228, | |
| "learning_rate": 0.0001825950061497276, | |
| "loss": 0.7983, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.13687580544983835, | |
| "learning_rate": 0.00018157090768283678, | |
| "loss": 0.8172, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.13705867781023107, | |
| "learning_rate": 0.00018052058181836151, | |
| "loss": 0.8212, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.15563306524282045, | |
| "learning_rate": 0.00017944436626633623, | |
| "loss": 0.8411, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.14396760217664073, | |
| "learning_rate": 0.00017834260706107595, | |
| "loss": 0.8108, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.12647869153054622, | |
| "learning_rate": 0.00017721565844991643, | |
| "loss": 0.8062, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.14745297927170234, | |
| "learning_rate": 0.00017606388277931328, | |
| "loss": 0.8744, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.15053882715392, | |
| "learning_rate": 0.0001748876503783373, | |
| "loss": 0.8352, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.1317924804547838, | |
| "learning_rate": 0.00017368733943960276, | |
| "loss": 0.8384, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.12819173037979142, | |
| "learning_rate": 0.00017246333589766787, | |
| "loss": 0.8531, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.1312708124132867, | |
| "learning_rate": 0.00017121603330494544, | |
| "loss": 0.8638, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.1411696833691384, | |
| "learning_rate": 0.0001699458327051647, | |
| "loss": 0.8382, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.14606044120399994, | |
| "learning_rate": 0.00016865314250442398, | |
| "loss": 0.8168, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.1455530152362454, | |
| "learning_rate": 0.00016733837833987633, | |
| "loss": 0.8117, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.14309897141049993, | |
| "learning_rate": 0.00016600196294609045, | |
| "loss": 0.8108, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.14312668748781052, | |
| "learning_rate": 0.00016464432601912912, | |
| "loss": 0.8303, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.12579071761666902, | |
| "learning_rate": 0.0001632659040783897, | |
| "loss": 0.8476, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.13350400452418196, | |
| "learning_rate": 0.00016186714032625035, | |
| "loss": 0.8526, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.14761760265503343, | |
| "learning_rate": 0.00016044848450556787, | |
| "loss": 0.8192, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.15245475954345805, | |
| "learning_rate": 0.00015901039275507245, | |
| "loss": 0.8298, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.15994496269289082, | |
| "learning_rate": 0.00015755332746270572, | |
| "loss": 0.8106, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.14107023644999184, | |
| "learning_rate": 0.00015607775711694977, | |
| "loss": 0.8378, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.15210143490634426, | |
| "learning_rate": 0.00015458415615619484, | |
| "loss": 0.8185, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 0.11955663423142318, | |
| "learning_rate": 0.00015307300481619333, | |
| "loss": 0.8576, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.13750521942283506, | |
| "learning_rate": 0.00015154478897565045, | |
| "loss": 0.881, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.11425649554717676, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.8139, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 0.1481189219653055, | |
| "learning_rate": 0.00014843913458341645, | |
| "loss": 0.8009, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 0.1588720467782321, | |
| "learning_rate": 0.00014686269458911332, | |
| "loss": 0.8177, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.14436853208913628, | |
| "learning_rate": 0.00014527118688797963, | |
| "loss": 0.9043, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.14039483442002895, | |
| "learning_rate": 0.0001436651231956064, | |
| "loss": 0.8292, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.1409577227905237, | |
| "learning_rate": 0.00014204501990775533, | |
| "loss": 0.8476, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.12770965177289617, | |
| "learning_rate": 0.00014041139793432274, | |
| "loss": 0.8266, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.13851032533153496, | |
| "learning_rate": 0.00013876478253185183, | |
| "loss": 0.8543, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.15009000078609738, | |
| "learning_rate": 0.00013710570313464778, | |
| "loss": 0.8273, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.1490899855893196, | |
| "learning_rate": 0.0001354346931845492, | |
| "loss": 0.8624, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.11669282581529886, | |
| "learning_rate": 0.00013375228995941133, | |
| "loss": 0.8476, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.12486593160211, | |
| "learning_rate": 0.0001320590344003557, | |
| "loss": 0.84, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.140864742842846, | |
| "learning_rate": 0.00013035547093784186, | |
| "loss": 0.8387, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 0.12552125440968517, | |
| "learning_rate": 0.00012864214731661742, | |
| "loss": 0.8177, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 0.14254214034385443, | |
| "learning_rate": 0.00012691961441960238, | |
| "loss": 0.8087, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.12161604299239723, | |
| "learning_rate": 0.00012518842609076413, | |
| "loss": 0.8255, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.14279805864279152, | |
| "learning_rate": 0.00012344913895704097, | |
| "loss": 0.8161, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.13061903554890092, | |
| "learning_rate": 0.00012170231224937032, | |
| "loss": 0.8101, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.14103602925499836, | |
| "learning_rate": 0.00011994850762287989, | |
| "loss": 0.8304, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.1437471461141528, | |
| "learning_rate": 0.0001181882889762994, | |
| "loss": 0.859, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.12804439910080107, | |
| "learning_rate": 0.00011642222227065089, | |
| "loss": 0.8103, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.16321939750119213, | |
| "learning_rate": 0.00011465087534727587, | |
| "loss": 0.8931, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.15577747479169646, | |
| "learning_rate": 0.0001128748177452581, | |
| "loss": 0.8272, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.13365546906519096, | |
| "learning_rate": 0.00011109462051830017, | |
| "loss": 0.8933, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.1290045452111647, | |
| "learning_rate": 0.00010931085605111354, | |
| "loss": 0.8053, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.13767998743232984, | |
| "learning_rate": 0.00010752409787538, | |
| "loss": 0.8352, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.1183523368788714, | |
| "learning_rate": 0.00010573492048534515, | |
| "loss": 0.7793, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.12680023137301807, | |
| "learning_rate": 0.00010394389915310149, | |
| "loss": 0.7862, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.1468717080766619, | |
| "learning_rate": 0.00010215160974362223, | |
| "loss": 0.8945, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.1340750536857251, | |
| "learning_rate": 0.00010035862852960387, | |
| "loss": 0.8677, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.1839710754750199, | |
| "learning_rate": 9.856553200617805e-05, | |
| "loss": 0.8004, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.13546348015830512, | |
| "learning_rate": 9.677289670555169e-05, | |
| "loss": 0.8346, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.13059522257925926, | |
| "learning_rate": 9.49812990116353e-05, | |
| "loss": 0.8545, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.1339705849021452, | |
| "learning_rate": 9.319131497471894e-05, | |
| "loss": 0.8575, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.11764292685223814, | |
| "learning_rate": 9.140352012625537e-05, | |
| "loss": 0.8284, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.13114039777564312, | |
| "learning_rate": 8.961848929381026e-05, | |
| "loss": 0.8509, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.13510854771345257, | |
| "learning_rate": 8.783679641623845e-05, | |
| "loss": 0.8578, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.15637793864418187, | |
| "learning_rate": 8.605901435914607e-05, | |
| "loss": 0.8106, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.12883059386856624, | |
| "learning_rate": 8.428571473069775e-05, | |
| "loss": 0.8364, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.1462312921265624, | |
| "learning_rate": 8.25174676978282e-05, | |
| "loss": 0.7912, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.14749765374112217, | |
| "learning_rate": 8.075484180291701e-05, | |
| "loss": 0.8091, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.12928243297252306, | |
| "learning_rate": 7.899840378098588e-05, | |
| "loss": 0.8456, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.13914653439360944, | |
| "learning_rate": 7.724871837747707e-05, | |
| "loss": 0.8543, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.14536772205298842, | |
| "learning_rate": 7.550634816667142e-05, | |
| "loss": 0.7879, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.16444943986829566, | |
| "learning_rate": 7.377185337080442e-05, | |
| "loss": 0.8295, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.15156151416294053, | |
| "learning_rate": 7.204579167993881e-05, | |
| "loss": 0.8328, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.11700169890112103, | |
| "learning_rate": 7.032871807265096e-05, | |
| "loss": 0.8549, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.14385917799771358, | |
| "learning_rate": 6.862118463758943e-05, | |
| "loss": 0.8116, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.12908350111171227, | |
| "learning_rate": 6.69237403959624e-05, | |
| "loss": 0.822, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.14851267167990456, | |
| "learning_rate": 6.52369311250116e-05, | |
| "loss": 0.8039, | |
| "step": 635 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.12710830010354282, | |
| "learning_rate": 6.356129918252927e-05, | |
| "loss": 0.8156, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.1425733932027878, | |
| "learning_rate": 6.189738333247432e-05, | |
| "loss": 0.8118, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 0.15619373129124595, | |
| "learning_rate": 6.024571857174443e-05, | |
| "loss": 0.8597, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 0.13569977800267843, | |
| "learning_rate": 5.860683595815893e-05, | |
| "loss": 0.8106, | |
| "step": 655 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.12464446293377444, | |
| "learning_rate": 5.698126243970845e-05, | |
| "loss": 0.8121, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.13451779337168288, | |
| "learning_rate": 5.536952068512608e-05, | |
| "loss": 0.795, | |
| "step": 665 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 0.12915761361111974, | |
| "learning_rate": 5.3772128915834184e-05, | |
| "loss": 0.9113, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 0.12809195814074534, | |
| "learning_rate": 5.218960073932122e-05, | |
| "loss": 0.8646, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.1278348440067524, | |
| "learning_rate": 5.062244498400228e-05, | |
| "loss": 0.8477, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.13317787598393002, | |
| "learning_rate": 4.907116553561607e-05, | |
| "loss": 0.8133, | |
| "step": 685 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.11667818988278275, | |
| "learning_rate": 4.753626117521103e-05, | |
| "loss": 0.7861, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.14046555978563427, | |
| "learning_rate": 4.601822541877291e-05, | |
| "loss": 0.7955, | |
| "step": 695 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.14732626535083013, | |
| "learning_rate": 4.451754635854517e-05, | |
| "loss": 0.8456, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.13263814935125381, | |
| "learning_rate": 4.303470650609325e-05, | |
| "loss": 0.8043, | |
| "step": 705 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 0.1360692214502702, | |
| "learning_rate": 4.1570182637163155e-05, | |
| "loss": 0.8857, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 0.1398967005930706, | |
| "learning_rate": 4.0124445638384366e-05, | |
| "loss": 0.7555, | |
| "step": 715 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.1408656242005034, | |
| "learning_rate": 3.869796035586625e-05, | |
| "loss": 0.8046, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.13431365357395633, | |
| "learning_rate": 3.7291185445736444e-05, | |
| "loss": 0.8074, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.12962057427444676, | |
| "learning_rate": 3.590457322666997e-05, | |
| "loss": 0.8509, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.1352493946183552, | |
| "learning_rate": 3.453856953445557e-05, | |
| "loss": 0.8446, | |
| "step": 735 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.11494147585476418, | |
| "learning_rate": 3.319361357864663e-05, | |
| "loss": 0.817, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.12332246306931878, | |
| "learning_rate": 3.187013780134291e-05, | |
| "loss": 0.843, | |
| "step": 745 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 0.13071315641894352, | |
| "learning_rate": 3.05685677381475e-05, | |
| "loss": 0.85, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 0.1382161424220368, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.8112, | |
| "step": 755 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 0.13955515168937047, | |
| "learning_rate": 2.8032811545345294e-05, | |
| "loss": 0.8608, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 0.1771167231447207, | |
| "learning_rate": 2.679944073443158e-05, | |
| "loss": 0.8045, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 0.14193864210665846, | |
| "learning_rate": 2.5589606012863963e-05, | |
| "loss": 0.8664, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.13840297934079954, | |
| "learning_rate": 2.4403696377371142e-05, | |
| "loss": 0.8663, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.13536851743943085, | |
| "learning_rate": 2.324209313207736e-05, | |
| "loss": 0.858, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.12646539460553902, | |
| "learning_rate": 2.210516976590179e-05, | |
| "loss": 0.8391, | |
| "step": 785 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.15367864492686661, | |
| "learning_rate": 2.099329183247126e-05, | |
| "loss": 0.8189, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.13297540559237797, | |
| "learning_rate": 1.9906816832584253e-05, | |
| "loss": 0.784, | |
| "step": 795 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.13099877023338075, | |
| "learning_rate": 1.8846094099263912e-05, | |
| "loss": 0.8318, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 0.14050104223112714, | |
| "learning_rate": 1.781146468543765e-05, | |
| "loss": 0.7681, | |
| "step": 805 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 0.1284097349682222, | |
| "learning_rate": 1.6803261254278636e-05, | |
| "loss": 0.8185, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.15940118109165669, | |
| "learning_rate": 1.582180797224507e-05, | |
| "loss": 0.8209, | |
| "step": 815 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.14361251131573077, | |
| "learning_rate": 1.4867420404851307e-05, | |
| "loss": 0.8281, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.1171536350981247, | |
| "learning_rate": 1.3940405415204416e-05, | |
| "loss": 0.8409, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.12744294436872022, | |
| "learning_rate": 1.30410610653389e-05, | |
| "loss": 0.8649, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.13903914731619177, | |
| "learning_rate": 1.2169676520381168e-05, | |
| "loss": 0.834, | |
| "step": 835 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.17232697048230508, | |
| "learning_rate": 1.1326531955574526e-05, | |
| "loss": 0.8537, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 0.1358777303761771, | |
| "learning_rate": 1.0511898466194903e-05, | |
| "loss": 0.8097, | |
| "step": 845 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 0.12398803923306301, | |
| "learning_rate": 9.726037980385738e-06, | |
| "loss": 0.7775, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.13477267750481708, | |
| "learning_rate": 8.969203174940654e-06, | |
| "loss": 0.8016, | |
| "step": 855 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.11581854986662919, | |
| "learning_rate": 8.24163739406062e-06, | |
| "loss": 0.8478, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.14736720617074886, | |
| "learning_rate": 7.543574571111655e-06, | |
| "loss": 0.844, | |
| "step": 865 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.14132183623868166, | |
| "learning_rate": 6.875239153408542e-06, | |
| "loss": 0.7973, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.1527267616049184, | |
| "learning_rate": 6.236846030048604e-06, | |
| "loss": 0.8063, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.12562228277842963, | |
| "learning_rate": 5.6286004628186675e-06, | |
| "loss": 0.8484, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 0.13601236333451983, | |
| "learning_rate": 5.0506980201973974e-06, | |
| "loss": 0.8285, | |
| "step": 885 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 0.145710784777074, | |
| "learning_rate": 4.503324514474483e-06, | |
| "loss": 0.8706, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.14032071629891896, | |
| "learning_rate": 3.986655942006579e-06, | |
| "loss": 0.8291, | |
| "step": 895 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.16354220706871617, | |
| "learning_rate": 3.5008584266294386e-06, | |
| "loss": 0.8065, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.14754573731673057, | |
| "learning_rate": 3.0460881662442763e-06, | |
| "loss": 0.872, | |
| "step": 905 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.13523433944792257, | |
| "learning_rate": 2.622491382595693e-06, | |
| "loss": 0.8479, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.13840943427588331, | |
| "learning_rate": 2.2302042742571193e-06, | |
| "loss": 0.8314, | |
| "step": 915 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.12888668453211496, | |
| "learning_rate": 1.869352972839067e-06, | |
| "loss": 0.8115, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.1620074090938547, | |
| "learning_rate": 1.5400535024342022e-06, | |
| "loss": 0.8037, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.15116632265694227, | |
| "learning_rate": 1.2424117423122328e-06, | |
| "loss": 0.8315, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.12857560053281283, | |
| "learning_rate": 9.765233928766493e-07, | |
| "loss": 0.7992, | |
| "step": 935 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.14514383958462998, | |
| "learning_rate": 7.42473944894384e-07, | |
| "loss": 0.8109, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.13037326046264752, | |
| "learning_rate": 5.403386520079323e-07, | |
| "loss": 0.7781, | |
| "step": 945 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.12930160005915564, | |
| "learning_rate": 3.701825065392184e-07, | |
| "loss": 0.8515, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 0.12470339296313583, | |
| "learning_rate": 2.320602185927001e-07, | |
| "loss": 0.7733, | |
| "step": 955 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.14117795257501795, | |
| "learning_rate": 1.2601619846444035e-07, | |
| "loss": 0.867, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.13083613982172085, | |
| "learning_rate": 5.208454236296234e-08, | |
| "loss": 0.8299, | |
| "step": 965 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.14645664746534057, | |
| "learning_rate": 1.0289021446308056e-08, | |
| "loss": 0.8721, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": NaN, | |
| "eval_runtime": 2192.2486, | |
| "eval_samples_per_second": 3.163, | |
| "eval_steps_per_second": 0.791, | |
| "step": 974 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 974, | |
| "total_flos": 1.053899276419072e+16, | |
| "train_loss": 0.8385560501527493, | |
| "train_runtime": 17932.7933, | |
| "train_samples_per_second": 3.477, | |
| "train_steps_per_second": 0.054 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 974, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "total_flos": 1.053899276419072e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |