| { | |
| "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": 0.41478540297292715, | |
| "learning_rate": 2.040816326530612e-06, | |
| "loss": 1.0591, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.5358244986842327, | |
| "learning_rate": 1.0204081632653061e-05, | |
| "loss": 0.9762, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.48168298126198883, | |
| "learning_rate": 2.0408163265306123e-05, | |
| "loss": 1.0246, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.23532241564777356, | |
| "learning_rate": 3.061224489795919e-05, | |
| "loss": 1.0114, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.2679701566666925, | |
| "learning_rate": 4.0816326530612245e-05, | |
| "loss": 0.9971, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.1583325918997123, | |
| "learning_rate": 5.102040816326531e-05, | |
| "loss": 0.9355, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.1840547641781333, | |
| "learning_rate": 6.122448979591838e-05, | |
| "loss": 0.943, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.239483711202957, | |
| "learning_rate": 7.142857142857143e-05, | |
| "loss": 0.8764, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.15385235740129172, | |
| "learning_rate": 8.163265306122449e-05, | |
| "loss": 0.8824, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.13852135103025864, | |
| "learning_rate": 9.183673469387756e-05, | |
| "loss": 0.8615, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.13036655403253458, | |
| "learning_rate": 0.00010204081632653062, | |
| "loss": 0.9202, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.138300443551494, | |
| "learning_rate": 0.00011224489795918367, | |
| "loss": 0.8849, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.12389157364724535, | |
| "learning_rate": 0.00012244897959183676, | |
| "loss": 0.862, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.1429134980556313, | |
| "learning_rate": 0.0001326530612244898, | |
| "loss": 0.8463, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.14541426965897125, | |
| "learning_rate": 0.00014285714285714287, | |
| "loss": 0.8289, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.15560066242946408, | |
| "learning_rate": 0.0001530612244897959, | |
| "loss": 0.8285, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.1382713303784162, | |
| "learning_rate": 0.00016326530612244898, | |
| "loss": 0.8267, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.1367658702482613, | |
| "learning_rate": 0.00017346938775510205, | |
| "loss": 0.8875, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.15684837460586795, | |
| "learning_rate": 0.00018367346938775512, | |
| "loss": 0.8758, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.13232147433779046, | |
| "learning_rate": 0.00019387755102040816, | |
| "loss": 0.8457, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.12839583748095243, | |
| "learning_rate": 0.0001999974277115551, | |
| "loss": 0.8492, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.1321027185844607, | |
| "learning_rate": 0.00019996849098629418, | |
| "loss": 0.787, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.13562231873522493, | |
| "learning_rate": 0.00019990741151022301, | |
| "loss": 0.8034, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.1872551577847746, | |
| "learning_rate": 0.0001998142089221534, | |
| "loss": 0.8897, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.11987471778945524, | |
| "learning_rate": 0.0001996889131894033, | |
| "loss": 0.8548, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.13758980449343652, | |
| "learning_rate": 0.00019953156459816179, | |
| "loss": 0.8126, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.12651848093312767, | |
| "learning_rate": 0.0001993422137405354, | |
| "loss": 0.8606, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.14411919368392379, | |
| "learning_rate": 0.00019912092149828174, | |
| "loss": 0.8858, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.150512988037676, | |
| "learning_rate": 0.00019886775902323405, | |
| "loss": 0.8593, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.15943254043396862, | |
| "learning_rate": 0.00019858280771442385, | |
| "loss": 0.8577, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.1934721467912684, | |
| "learning_rate": 0.00019826615919190887, | |
| "loss": 0.8628, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.1232771853573132, | |
| "learning_rate": 0.00019791791526731445, | |
| "loss": 0.84, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.14669498826369695, | |
| "learning_rate": 0.00019753818791109828, | |
| "loss": 0.8185, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.14085085759954724, | |
| "learning_rate": 0.0001971270992165486, | |
| "loss": 0.7795, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.1335331389930546, | |
| "learning_rate": 0.00019668478136052774, | |
| "loss": 0.8159, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.14301329004691762, | |
| "learning_rate": 0.0001962113765609735, | |
| "loss": 0.8008, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.1349013012428684, | |
| "learning_rate": 0.0001957070370311717, | |
| "loss": 0.8523, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.15345671624450827, | |
| "learning_rate": 0.00019517192493081565, | |
| "loss": 0.827, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.13105575744786446, | |
| "learning_rate": 0.00019460621231386676, | |
| "loss": 0.8844, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.14047662719322554, | |
| "learning_rate": 0.00019401008107323455, | |
| "loss": 0.8593, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.16716795882975244, | |
| "learning_rate": 0.0001933837228822925, | |
| "loss": 0.8105, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.14011283820384335, | |
| "learning_rate": 0.0001927273391332499, | |
| "loss": 0.8575, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.13652402697672666, | |
| "learning_rate": 0.00019204114087239806, | |
| "loss": 0.8615, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.1366384894487137, | |
| "learning_rate": 0.00019132534873225323, | |
| "loss": 0.8373, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.1418034030520275, | |
| "learning_rate": 0.00019058019286061665, | |
| "loss": 0.8335, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.12973981150967331, | |
| "learning_rate": 0.00018980591284657535, | |
| "loss": 0.8393, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.14098955432684662, | |
| "learning_rate": 0.00018900275764346768, | |
| "loss": 0.8324, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.1483857739019612, | |
| "learning_rate": 0.0001881709854888372, | |
| "loss": 0.8224, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.1207564764356093, | |
| "learning_rate": 0.00018731086382140226, | |
| "loss": 0.8115, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.147034450761397, | |
| "learning_rate": 0.00018642266919506644, | |
| "loss": 0.8473, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.16090713319886477, | |
| "learning_rate": 0.00018550668718999872, | |
| "loss": 0.7954, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.12088332198568695, | |
| "learning_rate": 0.0001845632123208111, | |
| "loss": 0.8553, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.13281843244805125, | |
| "learning_rate": 0.0001835925479418637, | |
| "loss": 0.8602, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.11395640254482223, | |
| "learning_rate": 0.0001825950061497276, | |
| "loss": 0.799, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.18338212518218974, | |
| "learning_rate": 0.00018157090768283678, | |
| "loss": 0.818, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.14005425053437323, | |
| "learning_rate": 0.00018052058181836151, | |
| "loss": 0.8221, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.15372527780279158, | |
| "learning_rate": 0.00017944436626633623, | |
| "loss": 0.8417, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.14434072633220638, | |
| "learning_rate": 0.00017834260706107595, | |
| "loss": 0.8108, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.12627644118944992, | |
| "learning_rate": 0.00017721565844991643, | |
| "loss": 0.8068, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.14772289738958688, | |
| "learning_rate": 0.00017606388277931328, | |
| "loss": 0.8744, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.13984571980575156, | |
| "learning_rate": 0.0001748876503783373, | |
| "loss": 0.835, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.13153373295044746, | |
| "learning_rate": 0.00017368733943960276, | |
| "loss": 0.8383, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.1291787835295784, | |
| "learning_rate": 0.00017246333589766787, | |
| "loss": 0.8532, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.1313754669175255, | |
| "learning_rate": 0.00017121603330494544, | |
| "loss": 0.8638, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.14687623249066953, | |
| "learning_rate": 0.0001699458327051647, | |
| "loss": 0.8385, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.14850598828123243, | |
| "learning_rate": 0.00016865314250442398, | |
| "loss": 0.8169, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.14865425552332714, | |
| "learning_rate": 0.00016733837833987633, | |
| "loss": 0.8129, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.14327447625468243, | |
| "learning_rate": 0.00016600196294609045, | |
| "loss": 0.8112, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.1422668164659587, | |
| "learning_rate": 0.00016464432601912912, | |
| "loss": 0.8306, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.12679383458857735, | |
| "learning_rate": 0.0001632659040783897, | |
| "loss": 0.8476, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.13298721576358818, | |
| "learning_rate": 0.00016186714032625035, | |
| "loss": 0.8527, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.15009708989503034, | |
| "learning_rate": 0.00016044848450556787, | |
| "loss": 0.819, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.1535277202050242, | |
| "learning_rate": 0.00015901039275507245, | |
| "loss": 0.8305, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.13563353212881551, | |
| "learning_rate": 0.00015755332746270572, | |
| "loss": 0.8104, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.14208140877531353, | |
| "learning_rate": 0.00015607775711694977, | |
| "loss": 0.8381, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.14084759655058215, | |
| "learning_rate": 0.00015458415615619484, | |
| "loss": 0.8186, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 0.1202290236098903, | |
| "learning_rate": 0.00015307300481619333, | |
| "loss": 0.8578, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.1427466626848493, | |
| "learning_rate": 0.00015154478897565045, | |
| "loss": 0.8816, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.11439109448818731, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.8142, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 0.1402674024104851, | |
| "learning_rate": 0.00014843913458341645, | |
| "loss": 0.8011, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 0.15497764154867305, | |
| "learning_rate": 0.00014686269458911332, | |
| "loss": 0.8179, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.14725740205615323, | |
| "learning_rate": 0.00014527118688797963, | |
| "loss": 0.9044, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.1441807780130091, | |
| "learning_rate": 0.0001436651231956064, | |
| "loss": 0.8295, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.14314161398501463, | |
| "learning_rate": 0.00014204501990775533, | |
| "loss": 0.8471, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.12813797598238083, | |
| "learning_rate": 0.00014041139793432274, | |
| "loss": 0.8262, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.14355138774702908, | |
| "learning_rate": 0.00013876478253185183, | |
| "loss": 0.8544, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.14065230597830594, | |
| "learning_rate": 0.00013710570313464778, | |
| "loss": 0.8273, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.1502376541192836, | |
| "learning_rate": 0.0001354346931845492, | |
| "loss": 0.8629, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.11612404557295304, | |
| "learning_rate": 0.00013375228995941133, | |
| "loss": 0.8474, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.12394764980545889, | |
| "learning_rate": 0.0001320590344003557, | |
| "loss": 0.8398, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.14219342684795552, | |
| "learning_rate": 0.00013035547093784186, | |
| "loss": 0.8386, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 0.13096131144311657, | |
| "learning_rate": 0.00012864214731661742, | |
| "loss": 0.8175, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 0.143251617809686, | |
| "learning_rate": 0.00012691961441960238, | |
| "loss": 0.8085, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.12498048690896839, | |
| "learning_rate": 0.00012518842609076413, | |
| "loss": 0.8259, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.14752810775417383, | |
| "learning_rate": 0.00012344913895704097, | |
| "loss": 0.8157, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.13076168345663483, | |
| "learning_rate": 0.00012170231224937032, | |
| "loss": 0.8102, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.14100703225317754, | |
| "learning_rate": 0.00011994850762287989, | |
| "loss": 0.8308, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.1406207969502619, | |
| "learning_rate": 0.0001181882889762994, | |
| "loss": 0.8538, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.12811621773423584, | |
| "learning_rate": 0.00011642222227065089, | |
| "loss": 0.8096, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.16240280156036038, | |
| "learning_rate": 0.00011465087534727587, | |
| "loss": 0.893, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.15744615260545805, | |
| "learning_rate": 0.0001128748177452581, | |
| "loss": 0.8271, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.13453736557228535, | |
| "learning_rate": 0.00011109462051830017, | |
| "loss": 0.8932, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.13488475951319145, | |
| "learning_rate": 0.00010931085605111354, | |
| "loss": 0.805, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.13615523867504728, | |
| "learning_rate": 0.00010752409787538, | |
| "loss": 0.8358, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.11608735125900543, | |
| "learning_rate": 0.00010573492048534515, | |
| "loss": 0.7792, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.12529530307366807, | |
| "learning_rate": 0.00010394389915310149, | |
| "loss": 0.7858, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.1477196524951594, | |
| "learning_rate": 0.00010215160974362223, | |
| "loss": 0.8949, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.13478280062013237, | |
| "learning_rate": 0.00010035862852960387, | |
| "loss": 0.8675, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.18751542446376887, | |
| "learning_rate": 9.856553200617805e-05, | |
| "loss": 0.8002, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.13384075191607198, | |
| "learning_rate": 9.677289670555169e-05, | |
| "loss": 0.8342, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.12830956280090938, | |
| "learning_rate": 9.49812990116353e-05, | |
| "loss": 0.8542, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.13139317688343669, | |
| "learning_rate": 9.319131497471894e-05, | |
| "loss": 0.857, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.11825939494831025, | |
| "learning_rate": 9.140352012625537e-05, | |
| "loss": 0.8286, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.1319552841228723, | |
| "learning_rate": 8.961848929381026e-05, | |
| "loss": 0.8513, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.13846831656086162, | |
| "learning_rate": 8.783679641623845e-05, | |
| "loss": 0.8575, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.1565132050678128, | |
| "learning_rate": 8.605901435914607e-05, | |
| "loss": 0.8095, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.1349070576898933, | |
| "learning_rate": 8.428571473069775e-05, | |
| "loss": 0.8364, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.1478930655811248, | |
| "learning_rate": 8.25174676978282e-05, | |
| "loss": 0.7913, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.14739768176324797, | |
| "learning_rate": 8.075484180291701e-05, | |
| "loss": 0.8095, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.13061570270501272, | |
| "learning_rate": 7.899840378098588e-05, | |
| "loss": 0.8456, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.1400435628778491, | |
| "learning_rate": 7.724871837747707e-05, | |
| "loss": 0.854, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.14619473905537722, | |
| "learning_rate": 7.550634816667142e-05, | |
| "loss": 0.7878, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.1669686546083547, | |
| "learning_rate": 7.377185337080442e-05, | |
| "loss": 0.8293, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.15161889018158878, | |
| "learning_rate": 7.204579167993881e-05, | |
| "loss": 0.8327, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.11802481456975004, | |
| "learning_rate": 7.032871807265096e-05, | |
| "loss": 0.8549, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.1428032348359374, | |
| "learning_rate": 6.862118463758943e-05, | |
| "loss": 0.8113, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.12867903817761486, | |
| "learning_rate": 6.69237403959624e-05, | |
| "loss": 0.8219, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.14771825500860036, | |
| "learning_rate": 6.52369311250116e-05, | |
| "loss": 0.8042, | |
| "step": 635 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.13258568806230905, | |
| "learning_rate": 6.356129918252927e-05, | |
| "loss": 0.8156, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.14732793338159728, | |
| "learning_rate": 6.189738333247432e-05, | |
| "loss": 0.8121, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 0.15966142035007472, | |
| "learning_rate": 6.024571857174443e-05, | |
| "loss": 0.8595, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 0.13684066043985474, | |
| "learning_rate": 5.860683595815893e-05, | |
| "loss": 0.8109, | |
| "step": 655 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.12490641045647621, | |
| "learning_rate": 5.698126243970845e-05, | |
| "loss": 0.8125, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.13460863241729326, | |
| "learning_rate": 5.536952068512608e-05, | |
| "loss": 0.7947, | |
| "step": 665 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 0.13010809900192385, | |
| "learning_rate": 5.3772128915834184e-05, | |
| "loss": 0.9054, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 0.12951084999951157, | |
| "learning_rate": 5.218960073932122e-05, | |
| "loss": 0.8647, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.12953418083912346, | |
| "learning_rate": 5.062244498400228e-05, | |
| "loss": 0.8474, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.13349564170084294, | |
| "learning_rate": 4.907116553561607e-05, | |
| "loss": 0.8127, | |
| "step": 685 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.11698335940312711, | |
| "learning_rate": 4.753626117521103e-05, | |
| "loss": 0.7862, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.1417412770863787, | |
| "learning_rate": 4.601822541877291e-05, | |
| "loss": 0.7952, | |
| "step": 695 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.1490752355077661, | |
| "learning_rate": 4.451754635854517e-05, | |
| "loss": 0.8458, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.1331101949433864, | |
| "learning_rate": 4.303470650609325e-05, | |
| "loss": 0.8044, | |
| "step": 705 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 0.13611270992724214, | |
| "learning_rate": 4.1570182637163155e-05, | |
| "loss": 0.8855, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 0.13998690568497368, | |
| "learning_rate": 4.0124445638384366e-05, | |
| "loss": 0.7553, | |
| "step": 715 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.14235394415159117, | |
| "learning_rate": 3.869796035586625e-05, | |
| "loss": 0.8048, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.13460415814884583, | |
| "learning_rate": 3.7291185445736444e-05, | |
| "loss": 0.808, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.13186297670947147, | |
| "learning_rate": 3.590457322666997e-05, | |
| "loss": 0.8512, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.13821443417948492, | |
| "learning_rate": 3.453856953445557e-05, | |
| "loss": 0.8445, | |
| "step": 735 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.11419625295174221, | |
| "learning_rate": 3.319361357864663e-05, | |
| "loss": 0.8169, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.12406541104741345, | |
| "learning_rate": 3.187013780134291e-05, | |
| "loss": 0.8428, | |
| "step": 745 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 0.1286809514714421, | |
| "learning_rate": 3.05685677381475e-05, | |
| "loss": 0.85, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 0.1445484766618607, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.8106, | |
| "step": 755 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 0.14262528146944597, | |
| "learning_rate": 2.8032811545345294e-05, | |
| "loss": 0.8609, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 0.1769607359536476, | |
| "learning_rate": 2.679944073443158e-05, | |
| "loss": 0.8045, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 0.14280821635903052, | |
| "learning_rate": 2.5589606012863963e-05, | |
| "loss": 0.866, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.1429520090865704, | |
| "learning_rate": 2.4403696377371142e-05, | |
| "loss": 0.8667, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.13554177395410102, | |
| "learning_rate": 2.324209313207736e-05, | |
| "loss": 0.8577, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.1270863540245396, | |
| "learning_rate": 2.210516976590179e-05, | |
| "loss": 0.839, | |
| "step": 785 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.15157612066040596, | |
| "learning_rate": 2.099329183247126e-05, | |
| "loss": 0.8193, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.13563544257174368, | |
| "learning_rate": 1.9906816832584253e-05, | |
| "loss": 0.7841, | |
| "step": 795 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.1438162068777128, | |
| "learning_rate": 1.8846094099263912e-05, | |
| "loss": 0.8317, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 0.14098563226328448, | |
| "learning_rate": 1.781146468543765e-05, | |
| "loss": 0.7682, | |
| "step": 805 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 0.12904056725011795, | |
| "learning_rate": 1.6803261254278636e-05, | |
| "loss": 0.8184, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.1577908132849086, | |
| "learning_rate": 1.582180797224507e-05, | |
| "loss": 0.8208, | |
| "step": 815 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.1448578052047199, | |
| "learning_rate": 1.4867420404851307e-05, | |
| "loss": 0.828, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.11696975769424958, | |
| "learning_rate": 1.3940405415204416e-05, | |
| "loss": 0.8405, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.12780753501046968, | |
| "learning_rate": 1.30410610653389e-05, | |
| "loss": 0.8647, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.1401641424372156, | |
| "learning_rate": 1.2169676520381168e-05, | |
| "loss": 0.8346, | |
| "step": 835 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.1507027680390559, | |
| "learning_rate": 1.1326531955574526e-05, | |
| "loss": 0.8532, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 0.13417895219140907, | |
| "learning_rate": 1.0511898466194903e-05, | |
| "loss": 0.8092, | |
| "step": 845 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 0.127954613918298, | |
| "learning_rate": 9.726037980385738e-06, | |
| "loss": 0.7773, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.1366880841170061, | |
| "learning_rate": 8.969203174940654e-06, | |
| "loss": 0.8018, | |
| "step": 855 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.11768872682875378, | |
| "learning_rate": 8.24163739406062e-06, | |
| "loss": 0.8483, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.1419611383110849, | |
| "learning_rate": 7.543574571111655e-06, | |
| "loss": 0.844, | |
| "step": 865 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.14220531250615795, | |
| "learning_rate": 6.875239153408542e-06, | |
| "loss": 0.7975, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.1566277092103407, | |
| "learning_rate": 6.236846030048604e-06, | |
| "loss": 0.8063, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.1268170098289612, | |
| "learning_rate": 5.6286004628186675e-06, | |
| "loss": 0.8484, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 0.13964640969590117, | |
| "learning_rate": 5.0506980201973974e-06, | |
| "loss": 0.8284, | |
| "step": 885 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 0.14682575127314904, | |
| "learning_rate": 4.503324514474483e-06, | |
| "loss": 0.8706, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.13996242963540317, | |
| "learning_rate": 3.986655942006579e-06, | |
| "loss": 0.8293, | |
| "step": 895 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.16776823076764466, | |
| "learning_rate": 3.5008584266294386e-06, | |
| "loss": 0.8063, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.1510152152640099, | |
| "learning_rate": 3.0460881662442763e-06, | |
| "loss": 0.8718, | |
| "step": 905 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.1366248738676145, | |
| "learning_rate": 2.622491382595693e-06, | |
| "loss": 0.8484, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.1404750930152043, | |
| "learning_rate": 2.2302042742571193e-06, | |
| "loss": 0.8313, | |
| "step": 915 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.1270118228269091, | |
| "learning_rate": 1.869352972839067e-06, | |
| "loss": 0.8117, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.16190747206009376, | |
| "learning_rate": 1.5400535024342022e-06, | |
| "loss": 0.8031, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.1529977790553585, | |
| "learning_rate": 1.2424117423122328e-06, | |
| "loss": 0.8316, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.12937443064788462, | |
| "learning_rate": 9.765233928766493e-07, | |
| "loss": 0.7992, | |
| "step": 935 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.1449373612680855, | |
| "learning_rate": 7.42473944894384e-07, | |
| "loss": 0.8107, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.12996651495057476, | |
| "learning_rate": 5.403386520079323e-07, | |
| "loss": 0.7778, | |
| "step": 945 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.12904475182773067, | |
| "learning_rate": 3.701825065392184e-07, | |
| "loss": 0.8515, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 0.12597952420128744, | |
| "learning_rate": 2.320602185927001e-07, | |
| "loss": 0.7737, | |
| "step": 955 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.1376254333598378, | |
| "learning_rate": 1.2601619846444035e-07, | |
| "loss": 0.8671, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.13114870010510168, | |
| "learning_rate": 5.208454236296234e-08, | |
| "loss": 0.8289, | |
| "step": 965 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.14712841990455128, | |
| "learning_rate": 1.0289021446308056e-08, | |
| "loss": 0.8719, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1840.0201, | |
| "eval_samples_per_second": 3.767, | |
| "eval_steps_per_second": 0.942, | |
| "step": 974 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 974, | |
| "total_flos": 1.052863723732992e+16, | |
| "train_loss": 0.838743479834445, | |
| "train_runtime": 18938.875, | |
| "train_samples_per_second": 3.293, | |
| "train_steps_per_second": 0.051 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 974, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "total_flos": 1.052863723732992e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |