| { | |
| "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": 0.278260000476033, | |
| "learning_rate": 2.040816326530612e-06, | |
| "loss": 0.8634, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.31777755687989134, | |
| "learning_rate": 1.0204081632653061e-05, | |
| "loss": 0.7908, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.2856436599152175, | |
| "learning_rate": 2.0408163265306123e-05, | |
| "loss": 0.8141, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.3147406591452229, | |
| "learning_rate": 3.061224489795919e-05, | |
| "loss": 0.8363, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.430777416418827, | |
| "learning_rate": 4.0816326530612245e-05, | |
| "loss": 0.79, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.25357137484315706, | |
| "learning_rate": 5.102040816326531e-05, | |
| "loss": 0.7292, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 0.2344949536176556, | |
| "learning_rate": 6.122448979591838e-05, | |
| "loss": 0.7751, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.25651285140456803, | |
| "learning_rate": 7.142857142857143e-05, | |
| "loss": 0.7369, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 0.2137957381062488, | |
| "learning_rate": 8.163265306122449e-05, | |
| "loss": 0.7419, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.20780118302244682, | |
| "learning_rate": 9.183673469387756e-05, | |
| "loss": 0.7366, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.2117698561394433, | |
| "learning_rate": 0.00010204081632653062, | |
| "loss": 0.7791, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.23271918782805745, | |
| "learning_rate": 0.00011224489795918367, | |
| "loss": 0.7575, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.2012595832299907, | |
| "learning_rate": 0.00012244897959183676, | |
| "loss": 0.7378, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.22236704776526806, | |
| "learning_rate": 0.0001326530612244898, | |
| "loss": 0.7239, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.2489335529197533, | |
| "learning_rate": 0.00014285714285714287, | |
| "loss": 0.7018, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.3158770567029429, | |
| "learning_rate": 0.0001530612244897959, | |
| "loss": 0.7198, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.22478841399771332, | |
| "learning_rate": 0.00016326530612244898, | |
| "loss": 0.6888, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.20279881782680315, | |
| "learning_rate": 0.00017346938775510205, | |
| "loss": 0.7614, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.25682553986207524, | |
| "learning_rate": 0.00018367346938775512, | |
| "loss": 0.7515, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.1933418650987126, | |
| "learning_rate": 0.00019387755102040816, | |
| "loss": 0.7114, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 0.19826076655023367, | |
| "learning_rate": 0.0001999974277115551, | |
| "loss": 0.7439, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.26988094840772753, | |
| "learning_rate": 0.00019996849098629418, | |
| "loss": 0.6878, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.2659652261451234, | |
| "learning_rate": 0.00019990741151022301, | |
| "loss": 0.7036, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.2673295115989153, | |
| "learning_rate": 0.0001998142089221534, | |
| "loss": 0.777, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.18585682140393867, | |
| "learning_rate": 0.0001996889131894033, | |
| "loss": 0.7384, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.20706654289391774, | |
| "learning_rate": 0.00019953156459816179, | |
| "loss": 0.7032, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.1975503949872139, | |
| "learning_rate": 0.0001993422137405354, | |
| "loss": 0.7394, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.19155990374714252, | |
| "learning_rate": 0.00019912092149828174, | |
| "loss": 0.7691, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.21896019533336944, | |
| "learning_rate": 0.00019886775902323405, | |
| "loss": 0.7513, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.21645922651700464, | |
| "learning_rate": 0.00019858280771442385, | |
| "loss": 0.7453, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.257201133497707, | |
| "learning_rate": 0.00019826615919190887, | |
| "loss": 0.7294, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.19592625320021334, | |
| "learning_rate": 0.00019791791526731445, | |
| "loss": 0.7338, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.22759246804125327, | |
| "learning_rate": 0.00019753818791109828, | |
| "loss": 0.7165, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.209587733155002, | |
| "learning_rate": 0.0001971270992165486, | |
| "loss": 0.6747, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.19532074422563744, | |
| "learning_rate": 0.00019668478136052774, | |
| "loss": 0.6955, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.2011921787107439, | |
| "learning_rate": 0.0001962113765609735, | |
| "loss": 0.6905, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.19355038708021588, | |
| "learning_rate": 0.0001957070370311717, | |
| "loss": 0.7285, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.2257845844487404, | |
| "learning_rate": 0.00019517192493081565, | |
| "loss": 0.7211, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.20633408505025505, | |
| "learning_rate": 0.00019460621231386676, | |
| "loss": 0.7662, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.19291580450963114, | |
| "learning_rate": 0.00019401008107323455, | |
| "loss": 0.7352, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.21208416809587324, | |
| "learning_rate": 0.0001933837228822925, | |
| "loss": 0.709, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.2138736486479049, | |
| "learning_rate": 0.0001927273391332499, | |
| "loss": 0.7385, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.1979519650921684, | |
| "learning_rate": 0.00019204114087239806, | |
| "loss": 0.7456, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.20036547859946416, | |
| "learning_rate": 0.00019132534873225323, | |
| "loss": 0.7314, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.2109051663933512, | |
| "learning_rate": 0.00019058019286061665, | |
| "loss": 0.7153, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.19663708727193688, | |
| "learning_rate": 0.00018980591284657535, | |
| "loss": 0.7367, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.22311526316960448, | |
| "learning_rate": 0.00018900275764346768, | |
| "loss": 0.7185, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.21856660380042922, | |
| "learning_rate": 0.0001881709854888372, | |
| "loss": 0.7106, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.1777614746714146, | |
| "learning_rate": 0.00018731086382140226, | |
| "loss": 0.7244, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.2152411939603674, | |
| "learning_rate": 0.00018642266919506644, | |
| "loss": 0.7278, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.21899542197544694, | |
| "learning_rate": 0.00018550668718999872, | |
| "loss": 0.6935, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.17879416628145775, | |
| "learning_rate": 0.0001845632123208111, | |
| "loss": 0.752, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.20502107907321068, | |
| "learning_rate": 0.0001835925479418637, | |
| "loss": 0.7524, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.1700894937720479, | |
| "learning_rate": 0.0001825950061497276, | |
| "loss": 0.6853, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.21767573335753063, | |
| "learning_rate": 0.00018157090768283678, | |
| "loss": 0.7021, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.19948224148229518, | |
| "learning_rate": 0.00018052058181836151, | |
| "loss": 0.7151, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.23643784041495308, | |
| "learning_rate": 0.00017944436626633623, | |
| "loss": 0.7341, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 0.22282846541984508, | |
| "learning_rate": 0.00017834260706107595, | |
| "loss": 0.704, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.19186041214173322, | |
| "learning_rate": 0.00017721565844991643, | |
| "loss": 0.6804, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.2308525508028877, | |
| "learning_rate": 0.00017606388277931328, | |
| "loss": 0.7458, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.19965653258936603, | |
| "learning_rate": 0.0001748876503783373, | |
| "loss": 0.7323, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.20597831398184133, | |
| "learning_rate": 0.00017368733943960276, | |
| "loss": 0.7271, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.2001333808257082, | |
| "learning_rate": 0.00017246333589766787, | |
| "loss": 0.745, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.20152879889137834, | |
| "learning_rate": 0.00017121603330494544, | |
| "loss": 0.7518, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.21123584799236225, | |
| "learning_rate": 0.0001699458327051647, | |
| "loss": 0.7354, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 0.21846335968696504, | |
| "learning_rate": 0.00016865314250442398, | |
| "loss": 0.7062, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.2167950261609826, | |
| "learning_rate": 0.00016733837833987633, | |
| "loss": 0.7031, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.22559549372774937, | |
| "learning_rate": 0.00016600196294609045, | |
| "loss": 0.7053, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.2054889277817906, | |
| "learning_rate": 0.00016464432601912912, | |
| "loss": 0.7177, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.18685675415965228, | |
| "learning_rate": 0.0001632659040783897, | |
| "loss": 0.7287, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.20347809434227057, | |
| "learning_rate": 0.00016186714032625035, | |
| "loss": 0.7409, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.23184228574982507, | |
| "learning_rate": 0.00016044848450556787, | |
| "loss": 0.7165, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.20860125908982083, | |
| "learning_rate": 0.00015901039275507245, | |
| "loss": 0.7103, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 0.20694892641136498, | |
| "learning_rate": 0.00015755332746270572, | |
| "loss": 0.7029, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.21942652565209622, | |
| "learning_rate": 0.00015607775711694977, | |
| "loss": 0.7146, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.2294923534717997, | |
| "learning_rate": 0.00015458415615619484, | |
| "loss": 0.7032, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 0.18361791504690167, | |
| "learning_rate": 0.00015307300481619333, | |
| "loss": 0.7416, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.21914429245426453, | |
| "learning_rate": 0.00015154478897565045, | |
| "loss": 0.7603, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.18463956028551098, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.7141, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 0.20739348085981676, | |
| "learning_rate": 0.00014843913458341645, | |
| "loss": 0.6928, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 0.23475300435624555, | |
| "learning_rate": 0.00014686269458911332, | |
| "loss": 0.7054, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.2051173887069507, | |
| "learning_rate": 0.00014527118688797963, | |
| "loss": 0.7507, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.24269848756324244, | |
| "learning_rate": 0.0001436651231956064, | |
| "loss": 0.7161, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.22304779640122963, | |
| "learning_rate": 0.00014204501990775533, | |
| "loss": 0.7338, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.2011370428613215, | |
| "learning_rate": 0.00014041139793432274, | |
| "loss": 0.7157, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.21389877949256977, | |
| "learning_rate": 0.00013876478253185183, | |
| "loss": 0.7332, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.2236862586451767, | |
| "learning_rate": 0.00013710570313464778, | |
| "loss": 0.7041, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.23099802308008413, | |
| "learning_rate": 0.0001354346931845492, | |
| "loss": 0.7497, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 0.19581351042893666, | |
| "learning_rate": 0.00013375228995941133, | |
| "loss": 0.7351, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.19487263901191804, | |
| "learning_rate": 0.0001320590344003557, | |
| "loss": 0.7308, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.21645686329510472, | |
| "learning_rate": 0.00013035547093784186, | |
| "loss": 0.7191, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 0.18394860544054228, | |
| "learning_rate": 0.00012864214731661742, | |
| "loss": 0.7057, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 0.22790900222471855, | |
| "learning_rate": 0.00012691961441960238, | |
| "loss": 0.6947, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.1987179503246446, | |
| "learning_rate": 0.00012518842609076413, | |
| "loss": 0.7185, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.21813756917111565, | |
| "learning_rate": 0.00012344913895704097, | |
| "loss": 0.729, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.20984163709630285, | |
| "learning_rate": 0.00012170231224937032, | |
| "loss": 0.7088, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.22260231112122383, | |
| "learning_rate": 0.00011994850762287989, | |
| "loss": 0.7286, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.20295712650274553, | |
| "learning_rate": 0.0001181882889762994, | |
| "loss": 0.7386, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.18927186766803714, | |
| "learning_rate": 0.00011642222227065089, | |
| "loss": 0.7075, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.23171766459875837, | |
| "learning_rate": 0.00011465087534727587, | |
| "loss": 0.7792, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.23250770712348723, | |
| "learning_rate": 0.0001128748177452581, | |
| "loss": 0.7122, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.19262951739342996, | |
| "learning_rate": 0.00011109462051830017, | |
| "loss": 0.7662, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.19092022364768874, | |
| "learning_rate": 0.00010931085605111354, | |
| "loss": 0.6973, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.21547485287745696, | |
| "learning_rate": 0.00010752409787538, | |
| "loss": 0.7269, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.19406360788612212, | |
| "learning_rate": 0.00010573492048534515, | |
| "loss": 0.6789, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.19763745138893807, | |
| "learning_rate": 0.00010394389915310149, | |
| "loss": 0.6799, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.23704613540002642, | |
| "learning_rate": 0.00010215160974362223, | |
| "loss": 0.7699, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.2077492206198066, | |
| "learning_rate": 0.00010035862852960387, | |
| "loss": 0.7494, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.24823136165061307, | |
| "learning_rate": 9.856553200617805e-05, | |
| "loss": 0.6884, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.2132059245507559, | |
| "learning_rate": 9.677289670555169e-05, | |
| "loss": 0.7244, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.19638919824879514, | |
| "learning_rate": 9.49812990116353e-05, | |
| "loss": 0.7417, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.20627175795252442, | |
| "learning_rate": 9.319131497471894e-05, | |
| "loss": 0.7505, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.18441794299779102, | |
| "learning_rate": 9.140352012625537e-05, | |
| "loss": 0.7232, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.21418665493484385, | |
| "learning_rate": 8.961848929381026e-05, | |
| "loss": 0.7378, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.203451633996103, | |
| "learning_rate": 8.783679641623845e-05, | |
| "loss": 0.7329, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.2320891827433096, | |
| "learning_rate": 8.605901435914607e-05, | |
| "loss": 0.7001, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.20083987402857975, | |
| "learning_rate": 8.428571473069775e-05, | |
| "loss": 0.7288, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.22859309749065204, | |
| "learning_rate": 8.25174676978282e-05, | |
| "loss": 0.6751, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.22056492293319443, | |
| "learning_rate": 8.075484180291701e-05, | |
| "loss": 0.6941, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.1948179419065092, | |
| "learning_rate": 7.899840378098588e-05, | |
| "loss": 0.739, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.21038078662707901, | |
| "learning_rate": 7.724871837747707e-05, | |
| "loss": 0.7376, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.21581379517101684, | |
| "learning_rate": 7.550634816667142e-05, | |
| "loss": 0.6979, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.23700658058033702, | |
| "learning_rate": 7.377185337080442e-05, | |
| "loss": 0.7074, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.22502925651665026, | |
| "learning_rate": 7.204579167993881e-05, | |
| "loss": 0.7299, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.18685509995796631, | |
| "learning_rate": 7.032871807265096e-05, | |
| "loss": 0.7474, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.1927226976109578, | |
| "learning_rate": 6.862118463758943e-05, | |
| "loss": 0.7128, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.1948509767269259, | |
| "learning_rate": 6.69237403959624e-05, | |
| "loss": 0.7161, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.2232281754415007, | |
| "learning_rate": 6.52369311250116e-05, | |
| "loss": 0.6987, | |
| "step": 635 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.18819418541155183, | |
| "learning_rate": 6.356129918252927e-05, | |
| "loss": 0.7199, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.21451153578042456, | |
| "learning_rate": 6.189738333247432e-05, | |
| "loss": 0.6883, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 0.21499927962963833, | |
| "learning_rate": 6.024571857174443e-05, | |
| "loss": 0.7393, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 0.21526007435620578, | |
| "learning_rate": 5.860683595815893e-05, | |
| "loss": 0.6947, | |
| "step": 655 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.18937695641617514, | |
| "learning_rate": 5.698126243970845e-05, | |
| "loss": 0.7021, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.20447447822914389, | |
| "learning_rate": 5.536952068512608e-05, | |
| "loss": 0.7083, | |
| "step": 665 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 0.19302906571345702, | |
| "learning_rate": 5.3772128915834184e-05, | |
| "loss": 0.7951, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 0.19472456937049842, | |
| "learning_rate": 5.218960073932122e-05, | |
| "loss": 0.7513, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.2098838114942542, | |
| "learning_rate": 5.062244498400228e-05, | |
| "loss": 0.7359, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.20138004255702624, | |
| "learning_rate": 4.907116553561607e-05, | |
| "loss": 0.7033, | |
| "step": 685 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.19181224343444572, | |
| "learning_rate": 4.753626117521103e-05, | |
| "loss": 0.6858, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.1949692307712896, | |
| "learning_rate": 4.601822541877291e-05, | |
| "loss": 0.6965, | |
| "step": 695 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.23054907760131058, | |
| "learning_rate": 4.451754635854517e-05, | |
| "loss": 0.7167, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.20250745838511075, | |
| "learning_rate": 4.303470650609325e-05, | |
| "loss": 0.6986, | |
| "step": 705 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 0.2046645255687334, | |
| "learning_rate": 4.1570182637163155e-05, | |
| "loss": 0.7575, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 0.2124511394223037, | |
| "learning_rate": 4.0124445638384366e-05, | |
| "loss": 0.6567, | |
| "step": 715 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.2135010011069348, | |
| "learning_rate": 3.869796035586625e-05, | |
| "loss": 0.7112, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.19119697720301349, | |
| "learning_rate": 3.7291185445736444e-05, | |
| "loss": 0.7012, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.2004872966103393, | |
| "learning_rate": 3.590457322666997e-05, | |
| "loss": 0.7411, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.21378384029632225, | |
| "learning_rate": 3.453856953445557e-05, | |
| "loss": 0.7387, | |
| "step": 735 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.17744782309443974, | |
| "learning_rate": 3.319361357864663e-05, | |
| "loss": 0.7101, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.17704977156913052, | |
| "learning_rate": 3.187013780134291e-05, | |
| "loss": 0.7288, | |
| "step": 745 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 0.18800266072214133, | |
| "learning_rate": 3.05685677381475e-05, | |
| "loss": 0.7495, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 0.21225887307816607, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.7122, | |
| "step": 755 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 0.25315618469033696, | |
| "learning_rate": 2.8032811545345294e-05, | |
| "loss": 0.7629, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 0.2514868870254598, | |
| "learning_rate": 2.679944073443158e-05, | |
| "loss": 0.7171, | |
| "step": 765 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 0.21366770249835562, | |
| "learning_rate": 2.5589606012863963e-05, | |
| "loss": 0.7391, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.22241099272695, | |
| "learning_rate": 2.4403696377371142e-05, | |
| "loss": 0.7448, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.20347299264526267, | |
| "learning_rate": 2.324209313207736e-05, | |
| "loss": 0.759, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.19865744808577407, | |
| "learning_rate": 2.210516976590179e-05, | |
| "loss": 0.7255, | |
| "step": 785 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 0.22352645971487345, | |
| "learning_rate": 2.099329183247126e-05, | |
| "loss": 0.717, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.19224761539756433, | |
| "learning_rate": 1.9906816832584253e-05, | |
| "loss": 0.6965, | |
| "step": 795 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.19123699782564682, | |
| "learning_rate": 1.8846094099263912e-05, | |
| "loss": 0.7251, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 0.20854016178992346, | |
| "learning_rate": 1.781146468543765e-05, | |
| "loss": 0.666, | |
| "step": 805 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 0.18983384744652834, | |
| "learning_rate": 1.6803261254278636e-05, | |
| "loss": 0.6989, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.24215867297829632, | |
| "learning_rate": 1.582180797224507e-05, | |
| "loss": 0.7072, | |
| "step": 815 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.20801086513570777, | |
| "learning_rate": 1.4867420404851307e-05, | |
| "loss": 0.7261, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.17293689072630952, | |
| "learning_rate": 1.3940405415204416e-05, | |
| "loss": 0.7297, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.19456327451562994, | |
| "learning_rate": 1.30410610653389e-05, | |
| "loss": 0.7584, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.22153981619648838, | |
| "learning_rate": 1.2169676520381168e-05, | |
| "loss": 0.72, | |
| "step": 835 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.21946612071940183, | |
| "learning_rate": 1.1326531955574526e-05, | |
| "loss": 0.7394, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 0.20107918472578915, | |
| "learning_rate": 1.0511898466194903e-05, | |
| "loss": 0.7165, | |
| "step": 845 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 0.19575920514235456, | |
| "learning_rate": 9.726037980385738e-06, | |
| "loss": 0.6878, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.21054264607284975, | |
| "learning_rate": 8.969203174940654e-06, | |
| "loss": 0.7001, | |
| "step": 855 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.18758969164604689, | |
| "learning_rate": 8.24163739406062e-06, | |
| "loss": 0.7333, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.1880041000902882, | |
| "learning_rate": 7.543574571111655e-06, | |
| "loss": 0.7388, | |
| "step": 865 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.21501638216282978, | |
| "learning_rate": 6.875239153408542e-06, | |
| "loss": 0.6992, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.22203256108917246, | |
| "learning_rate": 6.236846030048604e-06, | |
| "loss": 0.6998, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.18151009677487195, | |
| "learning_rate": 5.6286004628186675e-06, | |
| "loss": 0.7292, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 0.2069432233501742, | |
| "learning_rate": 5.0506980201973974e-06, | |
| "loss": 0.7304, | |
| "step": 885 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 0.20508701530744466, | |
| "learning_rate": 4.503324514474483e-06, | |
| "loss": 0.7542, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.20357657675306226, | |
| "learning_rate": 3.986655942006579e-06, | |
| "loss": 0.7251, | |
| "step": 895 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.2332850146858326, | |
| "learning_rate": 3.5008584266294386e-06, | |
| "loss": 0.7006, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.22950661568423028, | |
| "learning_rate": 3.0460881662442763e-06, | |
| "loss": 0.7516, | |
| "step": 905 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.21278393163752818, | |
| "learning_rate": 2.622491382595693e-06, | |
| "loss": 0.7212, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.20967762266666126, | |
| "learning_rate": 2.2302042742571193e-06, | |
| "loss": 0.7134, | |
| "step": 915 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.19675375943060863, | |
| "learning_rate": 1.869352972839067e-06, | |
| "loss": 0.7053, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.23714888815222485, | |
| "learning_rate": 1.5400535024342022e-06, | |
| "loss": 0.6874, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.2294513277463403, | |
| "learning_rate": 1.2424117423122328e-06, | |
| "loss": 0.7364, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.1903078496973518, | |
| "learning_rate": 9.765233928766493e-07, | |
| "loss": 0.7058, | |
| "step": 935 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.21650077932161488, | |
| "learning_rate": 7.42473944894384e-07, | |
| "loss": 0.6949, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.1884590964098191, | |
| "learning_rate": 5.403386520079323e-07, | |
| "loss": 0.681, | |
| "step": 945 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.19225795611194368, | |
| "learning_rate": 3.701825065392184e-07, | |
| "loss": 0.7427, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 0.18322950936594568, | |
| "learning_rate": 2.320602185927001e-07, | |
| "loss": 0.6803, | |
| "step": 955 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.2077977302422503, | |
| "learning_rate": 1.2601619846444035e-07, | |
| "loss": 0.7607, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.19247711440255688, | |
| "learning_rate": 5.208454236296234e-08, | |
| "loss": 0.7179, | |
| "step": 965 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.23115256209781324, | |
| "learning_rate": 1.0289021446308056e-08, | |
| "loss": 0.7473, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.74375319480896, | |
| "eval_runtime": 716.8033, | |
| "eval_samples_per_second": 9.672, | |
| "eval_steps_per_second": 0.605, | |
| "step": 974 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 974, | |
| "total_flos": 1.1285640269791232e+16, | |
| "train_loss": 0.724827410504069, | |
| "train_runtime": 16316.867, | |
| "train_samples_per_second": 3.822, | |
| "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": 1.1285640269791232e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |