| { | |
| "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": 0.2869611310957953, | |
| "learning_rate": 2.040816326530612e-06, | |
| "loss": 0.8612, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.3194590499756201, | |
| "learning_rate": 1.0204081632653061e-05, | |
| "loss": 0.7892, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.2900590043808106, | |
| "learning_rate": 2.0408163265306123e-05, | |
| "loss": 0.8123, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.32206451305877987, | |
| "learning_rate": 3.061224489795919e-05, | |
| "loss": 0.8338, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.412779025757005, | |
| "learning_rate": 4.0816326530612245e-05, | |
| "loss": 0.7852, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.2440444128385405, | |
| "learning_rate": 5.102040816326531e-05, | |
| "loss": 0.7278, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.23078040218918436, | |
| "learning_rate": 6.122448979591838e-05, | |
| "loss": 0.7749, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.2649545106643593, | |
| "learning_rate": 7.142857142857143e-05, | |
| "loss": 0.7369, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.21567429110623856, | |
| "learning_rate": 8.163265306122449e-05, | |
| "loss": 0.7412, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.21372272882412666, | |
| "learning_rate": 9.183673469387756e-05, | |
| "loss": 0.7375, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.2379221759632066, | |
| "learning_rate": 0.00010204081632653062, | |
| "loss": 0.7803, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.2382749273594966, | |
| "learning_rate": 0.00011224489795918367, | |
| "loss": 0.7585, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.20160910918787228, | |
| "learning_rate": 0.00012244897959183676, | |
| "loss": 0.7394, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.21140946440246422, | |
| "learning_rate": 0.0001326530612244898, | |
| "loss": 0.73, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.24165285545749063, | |
| "learning_rate": 0.00014285714285714287, | |
| "loss": 0.7034, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.26276753886817406, | |
| "learning_rate": 0.0001530612244897959, | |
| "loss": 0.7225, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.2278406577066701, | |
| "learning_rate": 0.00016326530612244898, | |
| "loss": 0.6887, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.20166487628632143, | |
| "learning_rate": 0.00017346938775510205, | |
| "loss": 0.7636, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.25404574731403684, | |
| "learning_rate": 0.00018367346938775512, | |
| "loss": 0.7536, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.20421230133266374, | |
| "learning_rate": 0.00019387755102040816, | |
| "loss": 0.7142, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.2106105346682244, | |
| "learning_rate": 0.0001999974277115551, | |
| "loss": 0.746, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.21445953241413981, | |
| "learning_rate": 0.00019996849098629418, | |
| "loss": 0.6897, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.21242787148744666, | |
| "learning_rate": 0.00019990741151022301, | |
| "loss": 0.7043, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.2897266739044473, | |
| "learning_rate": 0.0001998142089221534, | |
| "loss": 0.7799, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.1968138822308632, | |
| "learning_rate": 0.0001996889131894033, | |
| "loss": 0.7395, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.21497851174958627, | |
| "learning_rate": 0.00019953156459816179, | |
| "loss": 0.7059, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.21451890274852173, | |
| "learning_rate": 0.0001993422137405354, | |
| "loss": 0.7397, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.19014360167506802, | |
| "learning_rate": 0.00019912092149828174, | |
| "loss": 0.7686, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.22120779709504043, | |
| "learning_rate": 0.00019886775902323405, | |
| "loss": 0.7509, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.29083330661638745, | |
| "learning_rate": 0.00019858280771442385, | |
| "loss": 0.7443, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.27668275095195644, | |
| "learning_rate": 0.00019826615919190887, | |
| "loss": 0.7294, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.19750932915933125, | |
| "learning_rate": 0.00019791791526731445, | |
| "loss": 0.7338, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.22944621034884685, | |
| "learning_rate": 0.00019753818791109828, | |
| "loss": 0.7166, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.2109443098333667, | |
| "learning_rate": 0.0001971270992165486, | |
| "loss": 0.674, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.20676211281872686, | |
| "learning_rate": 0.00019668478136052774, | |
| "loss": 0.6949, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.20575242743242247, | |
| "learning_rate": 0.0001962113765609735, | |
| "loss": 0.6906, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.1951531711108696, | |
| "learning_rate": 0.0001957070370311717, | |
| "loss": 0.7289, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.22414853264275936, | |
| "learning_rate": 0.00019517192493081565, | |
| "loss": 0.7208, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.20668124787776296, | |
| "learning_rate": 0.00019460621231386676, | |
| "loss": 0.7671, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.19009017619221066, | |
| "learning_rate": 0.00019401008107323455, | |
| "loss": 0.7351, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.2189256627415435, | |
| "learning_rate": 0.0001933837228822925, | |
| "loss": 0.7093, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.2145651973089389, | |
| "learning_rate": 0.0001927273391332499, | |
| "loss": 0.7386, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.20465953071673096, | |
| "learning_rate": 0.00019204114087239806, | |
| "loss": 0.7456, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.20371247448101287, | |
| "learning_rate": 0.00019132534873225323, | |
| "loss": 0.7316, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.21926104230516402, | |
| "learning_rate": 0.00019058019286061665, | |
| "loss": 0.7153, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.21308006677872718, | |
| "learning_rate": 0.00018980591284657535, | |
| "loss": 0.7363, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.21830112014464587, | |
| "learning_rate": 0.00018900275764346768, | |
| "loss": 0.7188, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.22594321207444476, | |
| "learning_rate": 0.0001881709854888372, | |
| "loss": 0.7107, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.17764984572247824, | |
| "learning_rate": 0.00018731086382140226, | |
| "loss": 0.7247, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.21624903539690288, | |
| "learning_rate": 0.00018642266919506644, | |
| "loss": 0.7276, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.2171918383945808, | |
| "learning_rate": 0.00018550668718999872, | |
| "loss": 0.6937, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.17788572350886178, | |
| "learning_rate": 0.0001845632123208111, | |
| "loss": 0.7511, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.20199815612602928, | |
| "learning_rate": 0.0001835925479418637, | |
| "loss": 0.7522, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.173119607094577, | |
| "learning_rate": 0.0001825950061497276, | |
| "loss": 0.684, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.21342763081754199, | |
| "learning_rate": 0.00018157090768283678, | |
| "loss": 0.7026, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.19749144239575178, | |
| "learning_rate": 0.00018052058181836151, | |
| "loss": 0.7153, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.23666019704239125, | |
| "learning_rate": 0.00017944436626633623, | |
| "loss": 0.7337, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.22317738087512298, | |
| "learning_rate": 0.00017834260706107595, | |
| "loss": 0.7044, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.19219658466716574, | |
| "learning_rate": 0.00017721565844991643, | |
| "loss": 0.6799, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.23381913650905206, | |
| "learning_rate": 0.00017606388277931328, | |
| "loss": 0.7463, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.20207571127846147, | |
| "learning_rate": 0.0001748876503783373, | |
| "loss": 0.7305, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.20537100306933584, | |
| "learning_rate": 0.00017368733943960276, | |
| "loss": 0.7274, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.20177943742791943, | |
| "learning_rate": 0.00017246333589766787, | |
| "loss": 0.745, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.20364782308656443, | |
| "learning_rate": 0.00017121603330494544, | |
| "loss": 0.7519, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.21280082529891567, | |
| "learning_rate": 0.0001699458327051647, | |
| "loss": 0.736, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.2195869989993284, | |
| "learning_rate": 0.00016865314250442398, | |
| "loss": 0.7058, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.2179326990487266, | |
| "learning_rate": 0.00016733837833987633, | |
| "loss": 0.7033, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.22163995265105102, | |
| "learning_rate": 0.00016600196294609045, | |
| "loss": 0.7062, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.20563507571414916, | |
| "learning_rate": 0.00016464432601912912, | |
| "loss": 0.7177, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.1872905006878245, | |
| "learning_rate": 0.0001632659040783897, | |
| "loss": 0.7286, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.20516717514655824, | |
| "learning_rate": 0.00016186714032625035, | |
| "loss": 0.7409, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.22889695695378193, | |
| "learning_rate": 0.00016044848450556787, | |
| "loss": 0.7167, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.21338607450176578, | |
| "learning_rate": 0.00015901039275507245, | |
| "loss": 0.7118, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.20932529207974, | |
| "learning_rate": 0.00015755332746270572, | |
| "loss": 0.7026, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.21841220303292935, | |
| "learning_rate": 0.00015607775711694977, | |
| "loss": 0.7144, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.2249926395810304, | |
| "learning_rate": 0.00015458415615619484, | |
| "loss": 0.7036, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 0.18656241986804054, | |
| "learning_rate": 0.00015307300481619333, | |
| "loss": 0.7417, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.2157891411015049, | |
| "learning_rate": 0.00015154478897565045, | |
| "loss": 0.7602, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.18528004635067152, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.7143, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 0.20614234354483332, | |
| "learning_rate": 0.00014843913458341645, | |
| "loss": 0.693, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 0.23913126448103222, | |
| "learning_rate": 0.00014686269458911332, | |
| "loss": 0.7057, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.20395051777151454, | |
| "learning_rate": 0.00014527118688797963, | |
| "loss": 0.7512, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.24309987168907193, | |
| "learning_rate": 0.0001436651231956064, | |
| "loss": 0.7164, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.22288836478106547, | |
| "learning_rate": 0.00014204501990775533, | |
| "loss": 0.7333, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.19997689565338814, | |
| "learning_rate": 0.00014041139793432274, | |
| "loss": 0.716, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.21410827873496419, | |
| "learning_rate": 0.00013876478253185183, | |
| "loss": 0.7334, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.22446127836607146, | |
| "learning_rate": 0.00013710570313464778, | |
| "loss": 0.7042, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.23222688211452414, | |
| "learning_rate": 0.0001354346931845492, | |
| "loss": 0.7499, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.19295374405986063, | |
| "learning_rate": 0.00013375228995941133, | |
| "loss": 0.7357, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.19798750394457384, | |
| "learning_rate": 0.0001320590344003557, | |
| "loss": 0.7334, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.21722730155618872, | |
| "learning_rate": 0.00013035547093784186, | |
| "loss": 0.7193, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 0.184647192839392, | |
| "learning_rate": 0.00012864214731661742, | |
| "loss": 0.7062, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 0.2286779200568456, | |
| "learning_rate": 0.00012691961441960238, | |
| "loss": 0.6945, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.19896616453784802, | |
| "learning_rate": 0.00012518842609076413, | |
| "loss": 0.7188, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.21786128458289136, | |
| "learning_rate": 0.00012344913895704097, | |
| "loss": 0.7296, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.2070395610707047, | |
| "learning_rate": 0.00012170231224937032, | |
| "loss": 0.709, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.2267979628686904, | |
| "learning_rate": 0.00011994850762287989, | |
| "loss": 0.7283, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.20326820203968982, | |
| "learning_rate": 0.0001181882889762994, | |
| "loss": 0.739, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.19127608343670294, | |
| "learning_rate": 0.00011642222227065089, | |
| "loss": 0.7072, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.23023462990706073, | |
| "learning_rate": 0.00011465087534727587, | |
| "loss": 0.7792, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.2299774733193733, | |
| "learning_rate": 0.0001128748177452581, | |
| "loss": 0.7127, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.1933553970068975, | |
| "learning_rate": 0.00011109462051830017, | |
| "loss": 0.7673, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.19419056271814522, | |
| "learning_rate": 0.00010931085605111354, | |
| "loss": 0.6975, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.2142797681525271, | |
| "learning_rate": 0.00010752409787538, | |
| "loss": 0.7268, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.19406849444774402, | |
| "learning_rate": 0.00010573492048534515, | |
| "loss": 0.6817, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.19984584827497825, | |
| "learning_rate": 0.00010394389915310149, | |
| "loss": 0.6799, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.23853144293131298, | |
| "learning_rate": 0.00010215160974362223, | |
| "loss": 0.7705, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.20647647184488557, | |
| "learning_rate": 0.00010035862852960387, | |
| "loss": 0.7491, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.24615456413858594, | |
| "learning_rate": 9.856553200617805e-05, | |
| "loss": 0.6883, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.21240245143643857, | |
| "learning_rate": 9.677289670555169e-05, | |
| "loss": 0.7247, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.2000823201510655, | |
| "learning_rate": 9.49812990116353e-05, | |
| "loss": 0.7413, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.20666367113731635, | |
| "learning_rate": 9.319131497471894e-05, | |
| "loss": 0.75, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.18469259616208825, | |
| "learning_rate": 9.140352012625537e-05, | |
| "loss": 0.7214, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.21445560674239975, | |
| "learning_rate": 8.961848929381026e-05, | |
| "loss": 0.7376, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.2059315178687912, | |
| "learning_rate": 8.783679641623845e-05, | |
| "loss": 0.733, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.23365143820305953, | |
| "learning_rate": 8.605901435914607e-05, | |
| "loss": 0.7006, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.20038518152384194, | |
| "learning_rate": 8.428571473069775e-05, | |
| "loss": 0.7281, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.2270448364095405, | |
| "learning_rate": 8.25174676978282e-05, | |
| "loss": 0.6751, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.22128387907048663, | |
| "learning_rate": 8.075484180291701e-05, | |
| "loss": 0.694, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.1891101152811574, | |
| "learning_rate": 7.899840378098588e-05, | |
| "loss": 0.7384, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.21027626346886683, | |
| "learning_rate": 7.724871837747707e-05, | |
| "loss": 0.7372, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.22335920093565995, | |
| "learning_rate": 7.550634816667142e-05, | |
| "loss": 0.6979, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.23911224721224508, | |
| "learning_rate": 7.377185337080442e-05, | |
| "loss": 0.7079, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.22619894149171657, | |
| "learning_rate": 7.204579167993881e-05, | |
| "loss": 0.7306, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.18690561718768828, | |
| "learning_rate": 7.032871807265096e-05, | |
| "loss": 0.7476, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.19388179713391346, | |
| "learning_rate": 6.862118463758943e-05, | |
| "loss": 0.7121, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.19589622259075404, | |
| "learning_rate": 6.69237403959624e-05, | |
| "loss": 0.7156, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.22348889381112783, | |
| "learning_rate": 6.52369311250116e-05, | |
| "loss": 0.6988, | |
| "step": 635 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.19199107569507434, | |
| "learning_rate": 6.356129918252927e-05, | |
| "loss": 0.7195, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.21174035880989528, | |
| "learning_rate": 6.189738333247432e-05, | |
| "loss": 0.6889, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 0.21951618555807303, | |
| "learning_rate": 6.024571857174443e-05, | |
| "loss": 0.7401, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 0.21610930567187886, | |
| "learning_rate": 5.860683595815893e-05, | |
| "loss": 0.6946, | |
| "step": 655 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.1911939965176069, | |
| "learning_rate": 5.698126243970845e-05, | |
| "loss": 0.7026, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.20270357487417293, | |
| "learning_rate": 5.536952068512608e-05, | |
| "loss": 0.7083, | |
| "step": 665 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 0.19360058915764913, | |
| "learning_rate": 5.3772128915834184e-05, | |
| "loss": 0.7945, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 0.1977928757278191, | |
| "learning_rate": 5.218960073932122e-05, | |
| "loss": 0.7518, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.20960016035080886, | |
| "learning_rate": 5.062244498400228e-05, | |
| "loss": 0.7402, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.20026009581864332, | |
| "learning_rate": 4.907116553561607e-05, | |
| "loss": 0.7033, | |
| "step": 685 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.19145735043892229, | |
| "learning_rate": 4.753626117521103e-05, | |
| "loss": 0.6859, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.19584240345601217, | |
| "learning_rate": 4.601822541877291e-05, | |
| "loss": 0.6968, | |
| "step": 695 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.23270920498649578, | |
| "learning_rate": 4.451754635854517e-05, | |
| "loss": 0.7173, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.20792189933390842, | |
| "learning_rate": 4.303470650609325e-05, | |
| "loss": 0.6988, | |
| "step": 705 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 0.2063753862224208, | |
| "learning_rate": 4.1570182637163155e-05, | |
| "loss": 0.7576, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 0.21322675282501108, | |
| "learning_rate": 4.0124445638384366e-05, | |
| "loss": 0.6568, | |
| "step": 715 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.21703806382908236, | |
| "learning_rate": 3.869796035586625e-05, | |
| "loss": 0.711, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.1903818399050557, | |
| "learning_rate": 3.7291185445736444e-05, | |
| "loss": 0.7013, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.20064983895338837, | |
| "learning_rate": 3.590457322666997e-05, | |
| "loss": 0.741, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.21130061077032508, | |
| "learning_rate": 3.453856953445557e-05, | |
| "loss": 0.7383, | |
| "step": 735 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.17684598502717375, | |
| "learning_rate": 3.319361357864663e-05, | |
| "loss": 0.7096, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.17869261243212542, | |
| "learning_rate": 3.187013780134291e-05, | |
| "loss": 0.7285, | |
| "step": 745 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 0.18740684635500252, | |
| "learning_rate": 3.05685677381475e-05, | |
| "loss": 0.7475, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 0.21153099311303938, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.7122, | |
| "step": 755 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 0.26879324731183346, | |
| "learning_rate": 2.8032811545345294e-05, | |
| "loss": 0.7629, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 0.2477379977951793, | |
| "learning_rate": 2.679944073443158e-05, | |
| "loss": 0.7172, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 0.21621790135506005, | |
| "learning_rate": 2.5589606012863963e-05, | |
| "loss": 0.7397, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.22042903614192005, | |
| "learning_rate": 2.4403696377371142e-05, | |
| "loss": 0.7446, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.2046470002590343, | |
| "learning_rate": 2.324209313207736e-05, | |
| "loss": 0.7584, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.19842023146287918, | |
| "learning_rate": 2.210516976590179e-05, | |
| "loss": 0.7306, | |
| "step": 785 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.2271680029618541, | |
| "learning_rate": 2.099329183247126e-05, | |
| "loss": 0.7173, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.19284567420576632, | |
| "learning_rate": 1.9906816832584253e-05, | |
| "loss": 0.6972, | |
| "step": 795 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.18901685190843248, | |
| "learning_rate": 1.8846094099263912e-05, | |
| "loss": 0.7259, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 0.20493697097697364, | |
| "learning_rate": 1.781146468543765e-05, | |
| "loss": 0.6665, | |
| "step": 805 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 0.1892205656840172, | |
| "learning_rate": 1.6803261254278636e-05, | |
| "loss": 0.6993, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.24736999850815686, | |
| "learning_rate": 1.582180797224507e-05, | |
| "loss": 0.7075, | |
| "step": 815 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.21265134189137622, | |
| "learning_rate": 1.4867420404851307e-05, | |
| "loss": 0.7257, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.16998077993496108, | |
| "learning_rate": 1.3940405415204416e-05, | |
| "loss": 0.7293, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.19227746701204856, | |
| "learning_rate": 1.30410610653389e-05, | |
| "loss": 0.7578, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.2163980137011639, | |
| "learning_rate": 1.2169676520381168e-05, | |
| "loss": 0.7201, | |
| "step": 835 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.2174614380474134, | |
| "learning_rate": 1.1326531955574526e-05, | |
| "loss": 0.7393, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 0.19753139078286383, | |
| "learning_rate": 1.0511898466194903e-05, | |
| "loss": 0.7164, | |
| "step": 845 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 0.19652538273563885, | |
| "learning_rate": 9.726037980385738e-06, | |
| "loss": 0.687, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.20725040623745117, | |
| "learning_rate": 8.969203174940654e-06, | |
| "loss": 0.7006, | |
| "step": 855 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.1853922408804101, | |
| "learning_rate": 8.24163739406062e-06, | |
| "loss": 0.7334, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.18542913579514714, | |
| "learning_rate": 7.543574571111655e-06, | |
| "loss": 0.7387, | |
| "step": 865 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.2139733859772643, | |
| "learning_rate": 6.875239153408542e-06, | |
| "loss": 0.6985, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.22236312870883582, | |
| "learning_rate": 6.236846030048604e-06, | |
| "loss": 0.6998, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.1849060503719074, | |
| "learning_rate": 5.6286004628186675e-06, | |
| "loss": 0.7294, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 0.2090390382331485, | |
| "learning_rate": 5.0506980201973974e-06, | |
| "loss": 0.7304, | |
| "step": 885 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 0.2030892649120482, | |
| "learning_rate": 4.503324514474483e-06, | |
| "loss": 0.7548, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.20299856345700878, | |
| "learning_rate": 3.986655942006579e-06, | |
| "loss": 0.7247, | |
| "step": 895 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.2324930931474562, | |
| "learning_rate": 3.5008584266294386e-06, | |
| "loss": 0.7, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.22219104051624397, | |
| "learning_rate": 3.0460881662442763e-06, | |
| "loss": 0.7508, | |
| "step": 905 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.21326333904242326, | |
| "learning_rate": 2.622491382595693e-06, | |
| "loss": 0.7225, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.2098358511349563, | |
| "learning_rate": 2.2302042742571193e-06, | |
| "loss": 0.7132, | |
| "step": 915 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.1966565250666515, | |
| "learning_rate": 1.869352972839067e-06, | |
| "loss": 0.7059, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.2367264112088522, | |
| "learning_rate": 1.5400535024342022e-06, | |
| "loss": 0.6868, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.23055881813199589, | |
| "learning_rate": 1.2424117423122328e-06, | |
| "loss": 0.7367, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.1884312004939245, | |
| "learning_rate": 9.765233928766493e-07, | |
| "loss": 0.7055, | |
| "step": 935 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.21767998383781462, | |
| "learning_rate": 7.42473944894384e-07, | |
| "loss": 0.6955, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.18573736516667685, | |
| "learning_rate": 5.403386520079323e-07, | |
| "loss": 0.6811, | |
| "step": 945 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.1915354126552316, | |
| "learning_rate": 3.701825065392184e-07, | |
| "loss": 0.7429, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 0.1852158639659614, | |
| "learning_rate": 2.320602185927001e-07, | |
| "loss": 0.6799, | |
| "step": 955 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.21043187777240224, | |
| "learning_rate": 1.2601619846444035e-07, | |
| "loss": 0.7613, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.20739430949561452, | |
| "learning_rate": 5.208454236296234e-08, | |
| "loss": 0.7183, | |
| "step": 965 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.22836562458172804, | |
| "learning_rate": 1.0289021446308056e-08, | |
| "loss": 0.7469, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.7437335252761841, | |
| "eval_runtime": 726.8692, | |
| "eval_samples_per_second": 9.538, | |
| "eval_steps_per_second": 0.597, | |
| "step": 974 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 974, | |
| "total_flos": 1.1274999981146112e+16, | |
| "train_loss": 0.7250236611836255, | |
| "train_runtime": 16320.0317, | |
| "train_samples_per_second": 3.821, | |
| "train_steps_per_second": 0.06 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 974, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "total_flos": 1.1274999981146112e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |