llama3-sudo-5epochs-tofu_full_sft / trainer_state.json
Qin Liu
Model save
57fbd7c verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 5.0,
"eval_steps": 500,
"global_step": 1250,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.004,
"grad_norm": 176.61734163359293,
"learning_rate": 8e-08,
"loss": 4.3238,
"step": 1
},
{
"epoch": 0.02,
"grad_norm": 155.411055284858,
"learning_rate": 4.0000000000000003e-07,
"loss": 4.0646,
"step": 5
},
{
"epoch": 0.04,
"grad_norm": 133.32923635298718,
"learning_rate": 8.000000000000001e-07,
"loss": 3.906,
"step": 10
},
{
"epoch": 0.06,
"grad_norm": 80.21052252920832,
"learning_rate": 1.2000000000000002e-06,
"loss": 3.4641,
"step": 15
},
{
"epoch": 0.08,
"grad_norm": 48.85213712192445,
"learning_rate": 1.6000000000000001e-06,
"loss": 2.8464,
"step": 20
},
{
"epoch": 0.1,
"grad_norm": 45.776617949620864,
"learning_rate": 2.0000000000000003e-06,
"loss": 2.5666,
"step": 25
},
{
"epoch": 0.12,
"grad_norm": 26.73992232818656,
"learning_rate": 2.4000000000000003e-06,
"loss": 2.3056,
"step": 30
},
{
"epoch": 0.14,
"grad_norm": 17.729572998739787,
"learning_rate": 2.8000000000000003e-06,
"loss": 2.0261,
"step": 35
},
{
"epoch": 0.16,
"grad_norm": 15.223894174418577,
"learning_rate": 3.2000000000000003e-06,
"loss": 1.8306,
"step": 40
},
{
"epoch": 0.18,
"grad_norm": 17.850711162892512,
"learning_rate": 3.6000000000000003e-06,
"loss": 1.8819,
"step": 45
},
{
"epoch": 0.2,
"grad_norm": 23.333457209821216,
"learning_rate": 4.000000000000001e-06,
"loss": 1.7672,
"step": 50
},
{
"epoch": 0.22,
"grad_norm": 15.269663674152135,
"learning_rate": 4.4e-06,
"loss": 1.6371,
"step": 55
},
{
"epoch": 0.24,
"grad_norm": 17.5332417329875,
"learning_rate": 4.800000000000001e-06,
"loss": 1.561,
"step": 60
},
{
"epoch": 0.26,
"grad_norm": 17.643721758122435,
"learning_rate": 5.2e-06,
"loss": 1.5423,
"step": 65
},
{
"epoch": 0.28,
"grad_norm": 11.734090209997497,
"learning_rate": 5.600000000000001e-06,
"loss": 1.5258,
"step": 70
},
{
"epoch": 0.3,
"grad_norm": 12.798342916618239,
"learning_rate": 6e-06,
"loss": 1.4755,
"step": 75
},
{
"epoch": 0.32,
"grad_norm": 10.898202349824153,
"learning_rate": 6.4000000000000006e-06,
"loss": 1.5182,
"step": 80
},
{
"epoch": 0.34,
"grad_norm": 9.088500598331303,
"learning_rate": 6.800000000000001e-06,
"loss": 1.4451,
"step": 85
},
{
"epoch": 0.36,
"grad_norm": 10.983489966478087,
"learning_rate": 7.2000000000000005e-06,
"loss": 1.4034,
"step": 90
},
{
"epoch": 0.38,
"grad_norm": 10.83379202012167,
"learning_rate": 7.600000000000001e-06,
"loss": 1.425,
"step": 95
},
{
"epoch": 0.4,
"grad_norm": 8.672594081659222,
"learning_rate": 8.000000000000001e-06,
"loss": 1.3219,
"step": 100
},
{
"epoch": 0.42,
"grad_norm": 9.23377777855437,
"learning_rate": 8.400000000000001e-06,
"loss": 1.3799,
"step": 105
},
{
"epoch": 0.44,
"grad_norm": 9.166396489821778,
"learning_rate": 8.8e-06,
"loss": 1.4142,
"step": 110
},
{
"epoch": 0.46,
"grad_norm": 7.0660631261512545,
"learning_rate": 9.200000000000002e-06,
"loss": 1.417,
"step": 115
},
{
"epoch": 0.48,
"grad_norm": 8.099995733491154,
"learning_rate": 9.600000000000001e-06,
"loss": 1.4513,
"step": 120
},
{
"epoch": 0.5,
"grad_norm": 7.288660952017101,
"learning_rate": 1e-05,
"loss": 1.3968,
"step": 125
},
{
"epoch": 0.52,
"grad_norm": 11.534085489012515,
"learning_rate": 9.999512620046523e-06,
"loss": 1.3915,
"step": 130
},
{
"epoch": 0.54,
"grad_norm": 6.643837868783607,
"learning_rate": 9.998050575201772e-06,
"loss": 1.4451,
"step": 135
},
{
"epoch": 0.56,
"grad_norm": 8.528949559944818,
"learning_rate": 9.995614150494293e-06,
"loss": 1.3693,
"step": 140
},
{
"epoch": 0.58,
"grad_norm": 8.161397264280835,
"learning_rate": 9.992203820909906e-06,
"loss": 1.4197,
"step": 145
},
{
"epoch": 0.6,
"grad_norm": 7.882117733860753,
"learning_rate": 9.987820251299121e-06,
"loss": 1.3828,
"step": 150
},
{
"epoch": 0.62,
"grad_norm": 7.0280065330885995,
"learning_rate": 9.982464296247523e-06,
"loss": 1.4357,
"step": 155
},
{
"epoch": 0.64,
"grad_norm": 12.464091827730066,
"learning_rate": 9.976136999909156e-06,
"loss": 1.4414,
"step": 160
},
{
"epoch": 0.66,
"grad_norm": 6.4662291984046405,
"learning_rate": 9.968839595802982e-06,
"loss": 1.3905,
"step": 165
},
{
"epoch": 0.68,
"grad_norm": 6.65402151180169,
"learning_rate": 9.960573506572391e-06,
"loss": 1.4024,
"step": 170
},
{
"epoch": 0.7,
"grad_norm": 6.113615695740095,
"learning_rate": 9.951340343707852e-06,
"loss": 1.4059,
"step": 175
},
{
"epoch": 0.72,
"grad_norm": 6.665800966603425,
"learning_rate": 9.941141907232766e-06,
"loss": 1.331,
"step": 180
},
{
"epoch": 0.74,
"grad_norm": 6.685187563713334,
"learning_rate": 9.929980185352525e-06,
"loss": 1.3893,
"step": 185
},
{
"epoch": 0.76,
"grad_norm": 7.486719506086368,
"learning_rate": 9.91785735406693e-06,
"loss": 1.4275,
"step": 190
},
{
"epoch": 0.78,
"grad_norm": 5.888855898761242,
"learning_rate": 9.904775776745959e-06,
"loss": 1.4439,
"step": 195
},
{
"epoch": 0.8,
"grad_norm": 7.140808325295947,
"learning_rate": 9.890738003669029e-06,
"loss": 1.3542,
"step": 200
},
{
"epoch": 0.82,
"grad_norm": 6.336292620943448,
"learning_rate": 9.875746771527817e-06,
"loss": 1.413,
"step": 205
},
{
"epoch": 0.84,
"grad_norm": 7.079906024172797,
"learning_rate": 9.859805002892733e-06,
"loss": 1.3682,
"step": 210
},
{
"epoch": 0.86,
"grad_norm": 6.092306466503896,
"learning_rate": 9.842915805643156e-06,
"loss": 1.3372,
"step": 215
},
{
"epoch": 0.88,
"grad_norm": 7.149575331392461,
"learning_rate": 9.825082472361558e-06,
"loss": 1.3118,
"step": 220
},
{
"epoch": 0.9,
"grad_norm": 7.092971166660636,
"learning_rate": 9.806308479691595e-06,
"loss": 1.408,
"step": 225
},
{
"epoch": 0.92,
"grad_norm": 5.984139809585789,
"learning_rate": 9.786597487660336e-06,
"loss": 1.366,
"step": 230
},
{
"epoch": 0.94,
"grad_norm": 5.706031191925562,
"learning_rate": 9.765953338964736e-06,
"loss": 1.3484,
"step": 235
},
{
"epoch": 0.96,
"grad_norm": 6.2002602935958455,
"learning_rate": 9.744380058222483e-06,
"loss": 1.4072,
"step": 240
},
{
"epoch": 0.98,
"grad_norm": 5.8664026992140865,
"learning_rate": 9.721881851187406e-06,
"loss": 1.3566,
"step": 245
},
{
"epoch": 1.0,
"grad_norm": 5.9527848362443825,
"learning_rate": 9.698463103929542e-06,
"loss": 1.3774,
"step": 250
},
{
"epoch": 1.02,
"grad_norm": 6.374596828831512,
"learning_rate": 9.674128381980073e-06,
"loss": 0.9693,
"step": 255
},
{
"epoch": 1.04,
"grad_norm": 7.193932199691029,
"learning_rate": 9.648882429441258e-06,
"loss": 0.8766,
"step": 260
},
{
"epoch": 1.06,
"grad_norm": 6.524026562136591,
"learning_rate": 9.622730168061568e-06,
"loss": 0.9104,
"step": 265
},
{
"epoch": 1.08,
"grad_norm": 7.301263634828025,
"learning_rate": 9.595676696276173e-06,
"loss": 0.9334,
"step": 270
},
{
"epoch": 1.1,
"grad_norm": 5.670753086010248,
"learning_rate": 9.567727288213005e-06,
"loss": 0.8783,
"step": 275
},
{
"epoch": 1.12,
"grad_norm": 6.3479832974747525,
"learning_rate": 9.538887392664544e-06,
"loss": 0.9225,
"step": 280
},
{
"epoch": 1.1400000000000001,
"grad_norm": 5.625595616634227,
"learning_rate": 9.50916263202557e-06,
"loss": 0.9008,
"step": 285
},
{
"epoch": 1.16,
"grad_norm": 6.053431365139577,
"learning_rate": 9.478558801197065e-06,
"loss": 0.8751,
"step": 290
},
{
"epoch": 1.18,
"grad_norm": 6.445287713916751,
"learning_rate": 9.44708186645649e-06,
"loss": 0.8346,
"step": 295
},
{
"epoch": 1.2,
"grad_norm": 5.960574296608363,
"learning_rate": 9.414737964294636e-06,
"loss": 0.848,
"step": 300
},
{
"epoch": 1.22,
"grad_norm": 6.199669817245069,
"learning_rate": 9.381533400219319e-06,
"loss": 0.9026,
"step": 305
},
{
"epoch": 1.24,
"grad_norm": 6.03202099539365,
"learning_rate": 9.347474647526095e-06,
"loss": 0.9109,
"step": 310
},
{
"epoch": 1.26,
"grad_norm": 5.808928571078848,
"learning_rate": 9.312568346036288e-06,
"loss": 0.8967,
"step": 315
},
{
"epoch": 1.28,
"grad_norm": 5.49851518421863,
"learning_rate": 9.276821300802535e-06,
"loss": 0.8444,
"step": 320
},
{
"epoch": 1.3,
"grad_norm": 5.565049800849722,
"learning_rate": 9.24024048078213e-06,
"loss": 0.9639,
"step": 325
},
{
"epoch": 1.32,
"grad_norm": 5.562593169506186,
"learning_rate": 9.202833017478421e-06,
"loss": 0.8689,
"step": 330
},
{
"epoch": 1.34,
"grad_norm": 6.411205035883853,
"learning_rate": 9.164606203550498e-06,
"loss": 0.8717,
"step": 335
},
{
"epoch": 1.3599999999999999,
"grad_norm": 5.773518519345432,
"learning_rate": 9.125567491391476e-06,
"loss": 0.8878,
"step": 340
},
{
"epoch": 1.38,
"grad_norm": 6.106878743329819,
"learning_rate": 9.085724491675642e-06,
"loss": 0.8751,
"step": 345
},
{
"epoch": 1.4,
"grad_norm": 5.602954058571908,
"learning_rate": 9.045084971874738e-06,
"loss": 0.8821,
"step": 350
},
{
"epoch": 1.42,
"grad_norm": 5.241008563893136,
"learning_rate": 9.003656854743667e-06,
"loss": 0.9555,
"step": 355
},
{
"epoch": 1.44,
"grad_norm": 5.977919483177927,
"learning_rate": 8.961448216775955e-06,
"loss": 0.8809,
"step": 360
},
{
"epoch": 1.46,
"grad_norm": 5.666258321453365,
"learning_rate": 8.9184672866292e-06,
"loss": 0.8557,
"step": 365
},
{
"epoch": 1.48,
"grad_norm": 7.0549468534641235,
"learning_rate": 8.874722443520898e-06,
"loss": 0.9716,
"step": 370
},
{
"epoch": 1.5,
"grad_norm": 5.706625987649317,
"learning_rate": 8.83022221559489e-06,
"loss": 0.8705,
"step": 375
},
{
"epoch": 1.52,
"grad_norm": 6.104256802083156,
"learning_rate": 8.784975278258783e-06,
"loss": 0.8187,
"step": 380
},
{
"epoch": 1.54,
"grad_norm": 6.5871280297893415,
"learning_rate": 8.73899045249266e-06,
"loss": 0.9132,
"step": 385
},
{
"epoch": 1.56,
"grad_norm": 5.57782141266722,
"learning_rate": 8.692276703129421e-06,
"loss": 0.9218,
"step": 390
},
{
"epoch": 1.58,
"grad_norm": 5.472134179814514,
"learning_rate": 8.644843137107058e-06,
"loss": 0.8681,
"step": 395
},
{
"epoch": 1.6,
"grad_norm": 6.397480891094298,
"learning_rate": 8.596699001693257e-06,
"loss": 0.9391,
"step": 400
},
{
"epoch": 1.62,
"grad_norm": 5.363456755073,
"learning_rate": 8.547853682682605e-06,
"loss": 0.8772,
"step": 405
},
{
"epoch": 1.6400000000000001,
"grad_norm": 5.259511329956283,
"learning_rate": 8.498316702566828e-06,
"loss": 0.8644,
"step": 410
},
{
"epoch": 1.6600000000000001,
"grad_norm": 6.1682566237448775,
"learning_rate": 8.44809771867835e-06,
"loss": 0.947,
"step": 415
},
{
"epoch": 1.6800000000000002,
"grad_norm": 5.7859091344204145,
"learning_rate": 8.397206521307584e-06,
"loss": 0.9254,
"step": 420
},
{
"epoch": 1.7,
"grad_norm": 6.057814591654113,
"learning_rate": 8.345653031794292e-06,
"loss": 0.9062,
"step": 425
},
{
"epoch": 1.72,
"grad_norm": 5.465194913584291,
"learning_rate": 8.293447300593402e-06,
"loss": 0.9132,
"step": 430
},
{
"epoch": 1.74,
"grad_norm": 5.608611072004163,
"learning_rate": 8.240599505315656e-06,
"loss": 0.8963,
"step": 435
},
{
"epoch": 1.76,
"grad_norm": 6.514982052300432,
"learning_rate": 8.18711994874345e-06,
"loss": 0.9708,
"step": 440
},
{
"epoch": 1.78,
"grad_norm": 5.3243654718837075,
"learning_rate": 8.133019056822303e-06,
"loss": 0.8674,
"step": 445
},
{
"epoch": 1.8,
"grad_norm": 6.04340107619584,
"learning_rate": 8.078307376628292e-06,
"loss": 0.9128,
"step": 450
},
{
"epoch": 1.8199999999999998,
"grad_norm": 5.704899222715775,
"learning_rate": 8.022995574311876e-06,
"loss": 0.8828,
"step": 455
},
{
"epoch": 1.8399999999999999,
"grad_norm": 5.7674701231397805,
"learning_rate": 7.967094433018508e-06,
"loss": 0.9295,
"step": 460
},
{
"epoch": 1.8599999999999999,
"grad_norm": 5.324848982286405,
"learning_rate": 7.910614850786448e-06,
"loss": 0.8745,
"step": 465
},
{
"epoch": 1.88,
"grad_norm": 6.686861233685566,
"learning_rate": 7.85356783842216e-06,
"loss": 0.9646,
"step": 470
},
{
"epoch": 1.9,
"grad_norm": 6.121104493662131,
"learning_rate": 7.795964517353734e-06,
"loss": 0.933,
"step": 475
},
{
"epoch": 1.92,
"grad_norm": 5.582536228375127,
"learning_rate": 7.737816117462752e-06,
"loss": 0.9262,
"step": 480
},
{
"epoch": 1.94,
"grad_norm": 6.077848142278746,
"learning_rate": 7.679133974894984e-06,
"loss": 0.9029,
"step": 485
},
{
"epoch": 1.96,
"grad_norm": 5.986835723576084,
"learning_rate": 7.619929529850397e-06,
"loss": 0.9204,
"step": 490
},
{
"epoch": 1.98,
"grad_norm": 4.648465110523929,
"learning_rate": 7.560214324352858e-06,
"loss": 0.855,
"step": 495
},
{
"epoch": 2.0,
"grad_norm": 6.394027835473984,
"learning_rate": 7.500000000000001e-06,
"loss": 0.9621,
"step": 500
},
{
"epoch": 2.02,
"grad_norm": 3.89451985928487,
"learning_rate": 7.4392982956936644e-06,
"loss": 0.4544,
"step": 505
},
{
"epoch": 2.04,
"grad_norm": 6.182703543573361,
"learning_rate": 7.378121045351378e-06,
"loss": 0.4096,
"step": 510
},
{
"epoch": 2.06,
"grad_norm": 6.193574779151475,
"learning_rate": 7.31648017559931e-06,
"loss": 0.4193,
"step": 515
},
{
"epoch": 2.08,
"grad_norm": 4.8681830081803605,
"learning_rate": 7.254387703447154e-06,
"loss": 0.4205,
"step": 520
},
{
"epoch": 2.1,
"grad_norm": 5.210546829726289,
"learning_rate": 7.191855733945388e-06,
"loss": 0.4374,
"step": 525
},
{
"epoch": 2.12,
"grad_norm": 4.646814925775646,
"learning_rate": 7.128896457825364e-06,
"loss": 0.4225,
"step": 530
},
{
"epoch": 2.14,
"grad_norm": 4.654154374920671,
"learning_rate": 7.06552214912271e-06,
"loss": 0.4276,
"step": 535
},
{
"epoch": 2.16,
"grad_norm": 5.3434847628458275,
"learning_rate": 7.0017451627844765e-06,
"loss": 0.432,
"step": 540
},
{
"epoch": 2.18,
"grad_norm": 4.709540680953104,
"learning_rate": 6.9375779322605154e-06,
"loss": 0.4378,
"step": 545
},
{
"epoch": 2.2,
"grad_norm": 4.834968890809526,
"learning_rate": 6.873032967079562e-06,
"loss": 0.4405,
"step": 550
},
{
"epoch": 2.22,
"grad_norm": 4.21487141814997,
"learning_rate": 6.808122850410461e-06,
"loss": 0.4322,
"step": 555
},
{
"epoch": 2.24,
"grad_norm": 5.761355844826787,
"learning_rate": 6.7428602366090764e-06,
"loss": 0.4207,
"step": 560
},
{
"epoch": 2.26,
"grad_norm": 4.752918487437509,
"learning_rate": 6.677257848751276e-06,
"loss": 0.4514,
"step": 565
},
{
"epoch": 2.2800000000000002,
"grad_norm": 6.0722053004977585,
"learning_rate": 6.611328476152557e-06,
"loss": 0.4447,
"step": 570
},
{
"epoch": 2.3,
"grad_norm": 3.9317893606382217,
"learning_rate": 6.545084971874738e-06,
"loss": 0.447,
"step": 575
},
{
"epoch": 2.32,
"grad_norm": 4.6089986065133,
"learning_rate": 6.4785402502202345e-06,
"loss": 0.4267,
"step": 580
},
{
"epoch": 2.34,
"grad_norm": 5.5318341884342646,
"learning_rate": 6.411707284214384e-06,
"loss": 0.4495,
"step": 585
},
{
"epoch": 2.36,
"grad_norm": 6.3158633504097965,
"learning_rate": 6.344599103076329e-06,
"loss": 0.4502,
"step": 590
},
{
"epoch": 2.38,
"grad_norm": 4.522299833445048,
"learning_rate": 6.277228789678953e-06,
"loss": 0.4408,
"step": 595
},
{
"epoch": 2.4,
"grad_norm": 4.75003515089972,
"learning_rate": 6.209609477998339e-06,
"loss": 0.424,
"step": 600
},
{
"epoch": 2.42,
"grad_norm": 4.849076130585631,
"learning_rate": 6.141754350553279e-06,
"loss": 0.4418,
"step": 605
},
{
"epoch": 2.44,
"grad_norm": 4.754914150256052,
"learning_rate": 6.073676635835317e-06,
"loss": 0.4379,
"step": 610
},
{
"epoch": 2.46,
"grad_norm": 5.217922728452035,
"learning_rate": 6.005389605729824e-06,
"loss": 0.4223,
"step": 615
},
{
"epoch": 2.48,
"grad_norm": 5.09656877734741,
"learning_rate": 5.936906572928625e-06,
"loss": 0.4345,
"step": 620
},
{
"epoch": 2.5,
"grad_norm": 4.94943409663634,
"learning_rate": 5.8682408883346535e-06,
"loss": 0.4555,
"step": 625
},
{
"epoch": 2.52,
"grad_norm": 5.242233985299832,
"learning_rate": 5.799405938459175e-06,
"loss": 0.4513,
"step": 630
},
{
"epoch": 2.54,
"grad_norm": 4.4422684985992715,
"learning_rate": 5.730415142812059e-06,
"loss": 0.4288,
"step": 635
},
{
"epoch": 2.56,
"grad_norm": 3.912114397025565,
"learning_rate": 5.661281951285613e-06,
"loss": 0.4311,
"step": 640
},
{
"epoch": 2.58,
"grad_norm": 5.222721860673294,
"learning_rate": 5.592019841532507e-06,
"loss": 0.4648,
"step": 645
},
{
"epoch": 2.6,
"grad_norm": 4.724149942564251,
"learning_rate": 5.522642316338268e-06,
"loss": 0.4277,
"step": 650
},
{
"epoch": 2.62,
"grad_norm": 4.491400560249321,
"learning_rate": 5.453162900988902e-06,
"loss": 0.4369,
"step": 655
},
{
"epoch": 2.64,
"grad_norm": 4.8746449468292585,
"learning_rate": 5.383595140634093e-06,
"loss": 0.4079,
"step": 660
},
{
"epoch": 2.66,
"grad_norm": 3.82569667633887,
"learning_rate": 5.3139525976465675e-06,
"loss": 0.4441,
"step": 665
},
{
"epoch": 2.68,
"grad_norm": 4.346847846602506,
"learning_rate": 5.244248848978067e-06,
"loss": 0.45,
"step": 670
},
{
"epoch": 2.7,
"grad_norm": 4.388123844206694,
"learning_rate": 5.174497483512506e-06,
"loss": 0.4226,
"step": 675
},
{
"epoch": 2.7199999999999998,
"grad_norm": 4.9729250460950185,
"learning_rate": 5.1047120994167855e-06,
"loss": 0.4362,
"step": 680
},
{
"epoch": 2.74,
"grad_norm": 4.062911661829956,
"learning_rate": 5.034906301489808e-06,
"loss": 0.421,
"step": 685
},
{
"epoch": 2.76,
"grad_norm": 3.985815433151041,
"learning_rate": 4.965093698510192e-06,
"loss": 0.4436,
"step": 690
},
{
"epoch": 2.7800000000000002,
"grad_norm": 5.323464036561023,
"learning_rate": 4.895287900583216e-06,
"loss": 0.4503,
"step": 695
},
{
"epoch": 2.8,
"grad_norm": 5.374509098625144,
"learning_rate": 4.825502516487497e-06,
"loss": 0.4453,
"step": 700
},
{
"epoch": 2.82,
"grad_norm": 4.590575855627389,
"learning_rate": 4.755751151021934e-06,
"loss": 0.4284,
"step": 705
},
{
"epoch": 2.84,
"grad_norm": 4.509037906020891,
"learning_rate": 4.686047402353433e-06,
"loss": 0.4553,
"step": 710
},
{
"epoch": 2.86,
"grad_norm": 4.565251181592848,
"learning_rate": 4.6164048593659076e-06,
"loss": 0.4313,
"step": 715
},
{
"epoch": 2.88,
"grad_norm": 4.2913762971120155,
"learning_rate": 4.546837099011101e-06,
"loss": 0.4071,
"step": 720
},
{
"epoch": 2.9,
"grad_norm": 5.0125093823770905,
"learning_rate": 4.477357683661734e-06,
"loss": 0.4263,
"step": 725
},
{
"epoch": 2.92,
"grad_norm": 5.618873097905976,
"learning_rate": 4.4079801584674955e-06,
"loss": 0.4501,
"step": 730
},
{
"epoch": 2.94,
"grad_norm": 6.075104024673319,
"learning_rate": 4.3387180487143875e-06,
"loss": 0.4389,
"step": 735
},
{
"epoch": 2.96,
"grad_norm": 4.00631519669371,
"learning_rate": 4.269584857187942e-06,
"loss": 0.4373,
"step": 740
},
{
"epoch": 2.98,
"grad_norm": 4.705162020934786,
"learning_rate": 4.200594061540827e-06,
"loss": 0.4389,
"step": 745
},
{
"epoch": 3.0,
"grad_norm": 5.44214376188593,
"learning_rate": 4.131759111665349e-06,
"loss": 0.3925,
"step": 750
},
{
"epoch": 3.02,
"grad_norm": 2.7924019809148586,
"learning_rate": 4.063093427071376e-06,
"loss": 0.2656,
"step": 755
},
{
"epoch": 3.04,
"grad_norm": 3.104774247282136,
"learning_rate": 3.994610394270178e-06,
"loss": 0.2405,
"step": 760
},
{
"epoch": 3.06,
"grad_norm": 3.343609631464514,
"learning_rate": 3.926323364164684e-06,
"loss": 0.2467,
"step": 765
},
{
"epoch": 3.08,
"grad_norm": 3.419736245359143,
"learning_rate": 3.8582456494467214e-06,
"loss": 0.2723,
"step": 770
},
{
"epoch": 3.1,
"grad_norm": 3.8565429188392186,
"learning_rate": 3.790390522001662e-06,
"loss": 0.2657,
"step": 775
},
{
"epoch": 3.12,
"grad_norm": 4.254486576860473,
"learning_rate": 3.7227712103210485e-06,
"loss": 0.2665,
"step": 780
},
{
"epoch": 3.14,
"grad_norm": 3.665505016144709,
"learning_rate": 3.655400896923672e-06,
"loss": 0.2622,
"step": 785
},
{
"epoch": 3.16,
"grad_norm": 2.8301720206592416,
"learning_rate": 3.5882927157856175e-06,
"loss": 0.2619,
"step": 790
},
{
"epoch": 3.18,
"grad_norm": 3.693549461824111,
"learning_rate": 3.521459749779769e-06,
"loss": 0.2512,
"step": 795
},
{
"epoch": 3.2,
"grad_norm": 3.5206986483032723,
"learning_rate": 3.4549150281252635e-06,
"loss": 0.2526,
"step": 800
},
{
"epoch": 3.22,
"grad_norm": 3.3748884588505828,
"learning_rate": 3.3886715238474454e-06,
"loss": 0.2602,
"step": 805
},
{
"epoch": 3.24,
"grad_norm": 2.967732786836979,
"learning_rate": 3.322742151248726e-06,
"loss": 0.2606,
"step": 810
},
{
"epoch": 3.26,
"grad_norm": 4.073155951036809,
"learning_rate": 3.2571397633909252e-06,
"loss": 0.2602,
"step": 815
},
{
"epoch": 3.2800000000000002,
"grad_norm": 3.1633808640117236,
"learning_rate": 3.1918771495895395e-06,
"loss": 0.2455,
"step": 820
},
{
"epoch": 3.3,
"grad_norm": 3.1430107416422763,
"learning_rate": 3.12696703292044e-06,
"loss": 0.2532,
"step": 825
},
{
"epoch": 3.32,
"grad_norm": 4.278015207900964,
"learning_rate": 3.0624220677394854e-06,
"loss": 0.2491,
"step": 830
},
{
"epoch": 3.34,
"grad_norm": 4.508283169888662,
"learning_rate": 2.9982548372155264e-06,
"loss": 0.2698,
"step": 835
},
{
"epoch": 3.36,
"grad_norm": 3.3985357213270353,
"learning_rate": 2.934477850877292e-06,
"loss": 0.271,
"step": 840
},
{
"epoch": 3.38,
"grad_norm": 4.3177739641475625,
"learning_rate": 2.871103542174637e-06,
"loss": 0.2723,
"step": 845
},
{
"epoch": 3.4,
"grad_norm": 3.6904622227059964,
"learning_rate": 2.8081442660546126e-06,
"loss": 0.2593,
"step": 850
},
{
"epoch": 3.42,
"grad_norm": 2.9293826019569074,
"learning_rate": 2.7456122965528475e-06,
"loss": 0.2617,
"step": 855
},
{
"epoch": 3.44,
"grad_norm": 3.290303151322462,
"learning_rate": 2.683519824400693e-06,
"loss": 0.2582,
"step": 860
},
{
"epoch": 3.46,
"grad_norm": 3.2377051712154716,
"learning_rate": 2.6218789546486235e-06,
"loss": 0.262,
"step": 865
},
{
"epoch": 3.48,
"grad_norm": 3.968934702000483,
"learning_rate": 2.560701704306336e-06,
"loss": 0.2527,
"step": 870
},
{
"epoch": 3.5,
"grad_norm": 2.814982968549524,
"learning_rate": 2.5000000000000015e-06,
"loss": 0.2547,
"step": 875
},
{
"epoch": 3.52,
"grad_norm": 2.8738241230711576,
"learning_rate": 2.4397856756471435e-06,
"loss": 0.2584,
"step": 880
},
{
"epoch": 3.54,
"grad_norm": 3.2477024834709396,
"learning_rate": 2.380070470149605e-06,
"loss": 0.2514,
"step": 885
},
{
"epoch": 3.56,
"grad_norm": 4.0745774880578844,
"learning_rate": 2.320866025105016e-06,
"loss": 0.266,
"step": 890
},
{
"epoch": 3.58,
"grad_norm": 3.440405002018834,
"learning_rate": 2.2621838825372496e-06,
"loss": 0.2637,
"step": 895
},
{
"epoch": 3.6,
"grad_norm": 3.3065115492957933,
"learning_rate": 2.204035482646267e-06,
"loss": 0.2589,
"step": 900
},
{
"epoch": 3.62,
"grad_norm": 2.495985385885236,
"learning_rate": 2.146432161577842e-06,
"loss": 0.2552,
"step": 905
},
{
"epoch": 3.64,
"grad_norm": 2.9350609154945473,
"learning_rate": 2.0893851492135536e-06,
"loss": 0.2549,
"step": 910
},
{
"epoch": 3.66,
"grad_norm": 3.6046449037840222,
"learning_rate": 2.0329055669814936e-06,
"loss": 0.2591,
"step": 915
},
{
"epoch": 3.68,
"grad_norm": 3.5165798523906577,
"learning_rate": 1.977004425688126e-06,
"loss": 0.2616,
"step": 920
},
{
"epoch": 3.7,
"grad_norm": 3.126960853099222,
"learning_rate": 1.9216926233717087e-06,
"loss": 0.2582,
"step": 925
},
{
"epoch": 3.7199999999999998,
"grad_norm": 3.0303889926026693,
"learning_rate": 1.8669809431776991e-06,
"loss": 0.2414,
"step": 930
},
{
"epoch": 3.74,
"grad_norm": 3.4976844042391506,
"learning_rate": 1.8128800512565514e-06,
"loss": 0.2654,
"step": 935
},
{
"epoch": 3.76,
"grad_norm": 4.624711265706133,
"learning_rate": 1.7594004946843458e-06,
"loss": 0.2601,
"step": 940
},
{
"epoch": 3.7800000000000002,
"grad_norm": 3.1163969659292547,
"learning_rate": 1.7065526994065973e-06,
"loss": 0.2635,
"step": 945
},
{
"epoch": 3.8,
"grad_norm": 4.134091674876607,
"learning_rate": 1.6543469682057105e-06,
"loss": 0.2613,
"step": 950
},
{
"epoch": 3.82,
"grad_norm": 4.520449029040607,
"learning_rate": 1.6027934786924187e-06,
"loss": 0.2659,
"step": 955
},
{
"epoch": 3.84,
"grad_norm": 3.340691260013119,
"learning_rate": 1.551902281321651e-06,
"loss": 0.263,
"step": 960
},
{
"epoch": 3.86,
"grad_norm": 3.3324205347544362,
"learning_rate": 1.5016832974331725e-06,
"loss": 0.2674,
"step": 965
},
{
"epoch": 3.88,
"grad_norm": 3.1997988068352248,
"learning_rate": 1.4521463173173966e-06,
"loss": 0.25,
"step": 970
},
{
"epoch": 3.9,
"grad_norm": 2.4626614462088248,
"learning_rate": 1.4033009983067454e-06,
"loss": 0.2622,
"step": 975
},
{
"epoch": 3.92,
"grad_norm": 2.5755921562893342,
"learning_rate": 1.3551568628929434e-06,
"loss": 0.246,
"step": 980
},
{
"epoch": 3.94,
"grad_norm": 2.9337304572329717,
"learning_rate": 1.3077232968705805e-06,
"loss": 0.2403,
"step": 985
},
{
"epoch": 3.96,
"grad_norm": 3.6309216538671354,
"learning_rate": 1.2610095475073415e-06,
"loss": 0.2708,
"step": 990
},
{
"epoch": 3.98,
"grad_norm": 2.6536626242412056,
"learning_rate": 1.2150247217412186e-06,
"loss": 0.2495,
"step": 995
},
{
"epoch": 4.0,
"grad_norm": 3.8919822758192404,
"learning_rate": 1.1697777844051105e-06,
"loss": 0.2763,
"step": 1000
},
{
"epoch": 4.02,
"grad_norm": 2.6296022264068797,
"learning_rate": 1.1252775564791023e-06,
"loss": 0.2009,
"step": 1005
},
{
"epoch": 4.04,
"grad_norm": 2.0687616816587715,
"learning_rate": 1.0815327133708015e-06,
"loss": 0.1978,
"step": 1010
},
{
"epoch": 4.06,
"grad_norm": 2.036715506169882,
"learning_rate": 1.0385517832240472e-06,
"loss": 0.195,
"step": 1015
},
{
"epoch": 4.08,
"grad_norm": 2.1444679630444647,
"learning_rate": 9.963431452563331e-07,
"loss": 0.195,
"step": 1020
},
{
"epoch": 4.1,
"grad_norm": 2.3208387437528306,
"learning_rate": 9.549150281252633e-07,
"loss": 0.2003,
"step": 1025
},
{
"epoch": 4.12,
"grad_norm": 2.6475787949798635,
"learning_rate": 9.142755083243577e-07,
"loss": 0.2012,
"step": 1030
},
{
"epoch": 4.14,
"grad_norm": 2.0747944566043417,
"learning_rate": 8.744325086085248e-07,
"loss": 0.1839,
"step": 1035
},
{
"epoch": 4.16,
"grad_norm": 2.5790204025275174,
"learning_rate": 8.353937964495029e-07,
"loss": 0.2078,
"step": 1040
},
{
"epoch": 4.18,
"grad_norm": 2.279012511901917,
"learning_rate": 7.971669825215789e-07,
"loss": 0.2033,
"step": 1045
},
{
"epoch": 4.2,
"grad_norm": 2.263082794933402,
"learning_rate": 7.597595192178702e-07,
"loss": 0.2013,
"step": 1050
},
{
"epoch": 4.22,
"grad_norm": 1.746174251156355,
"learning_rate": 7.23178699197467e-07,
"loss": 0.1942,
"step": 1055
},
{
"epoch": 4.24,
"grad_norm": 1.8889622200380844,
"learning_rate": 6.874316539637127e-07,
"loss": 0.192,
"step": 1060
},
{
"epoch": 4.26,
"grad_norm": 1.9542919003413253,
"learning_rate": 6.52525352473905e-07,
"loss": 0.2073,
"step": 1065
},
{
"epoch": 4.28,
"grad_norm": 2.110295600169004,
"learning_rate": 6.184665997806832e-07,
"loss": 0.1926,
"step": 1070
},
{
"epoch": 4.3,
"grad_norm": 2.2234924989126563,
"learning_rate": 5.852620357053651e-07,
"loss": 0.2033,
"step": 1075
},
{
"epoch": 4.32,
"grad_norm": 2.6413829746557504,
"learning_rate": 5.529181335435124e-07,
"loss": 0.198,
"step": 1080
},
{
"epoch": 4.34,
"grad_norm": 2.097744663828809,
"learning_rate": 5.214411988029355e-07,
"loss": 0.2017,
"step": 1085
},
{
"epoch": 4.36,
"grad_norm": 2.3107650980894263,
"learning_rate": 4.908373679744316e-07,
"loss": 0.2004,
"step": 1090
},
{
"epoch": 4.38,
"grad_norm": 2.0392932247609665,
"learning_rate": 4.6111260733545714e-07,
"loss": 0.1978,
"step": 1095
},
{
"epoch": 4.4,
"grad_norm": 1.9108697879670782,
"learning_rate": 4.322727117869951e-07,
"loss": 0.2064,
"step": 1100
},
{
"epoch": 4.42,
"grad_norm": 2.0940301187428645,
"learning_rate": 4.043233037238281e-07,
"loss": 0.1992,
"step": 1105
},
{
"epoch": 4.44,
"grad_norm": 2.1573212002556192,
"learning_rate": 3.772698319384349e-07,
"loss": 0.1978,
"step": 1110
},
{
"epoch": 4.46,
"grad_norm": 1.7681339799095577,
"learning_rate": 3.511175705587433e-07,
"loss": 0.1949,
"step": 1115
},
{
"epoch": 4.48,
"grad_norm": 2.6507223455216335,
"learning_rate": 3.258716180199278e-07,
"loss": 0.1974,
"step": 1120
},
{
"epoch": 4.5,
"grad_norm": 4.899195247798816,
"learning_rate": 3.015368960704584e-07,
"loss": 0.1951,
"step": 1125
},
{
"epoch": 4.52,
"grad_norm": 2.362208067582404,
"learning_rate": 2.7811814881259503e-07,
"loss": 0.1898,
"step": 1130
},
{
"epoch": 4.54,
"grad_norm": 1.9828163543826154,
"learning_rate": 2.556199417775174e-07,
"loss": 0.2081,
"step": 1135
},
{
"epoch": 4.5600000000000005,
"grad_norm": 1.9187249803005326,
"learning_rate": 2.3404666103526542e-07,
"loss": 0.2027,
"step": 1140
},
{
"epoch": 4.58,
"grad_norm": 2.050797728763808,
"learning_rate": 2.134025123396638e-07,
"loss": 0.195,
"step": 1145
},
{
"epoch": 4.6,
"grad_norm": 2.434262719508876,
"learning_rate": 1.9369152030840553e-07,
"loss": 0.1935,
"step": 1150
},
{
"epoch": 4.62,
"grad_norm": 2.7114986108898496,
"learning_rate": 1.7491752763844294e-07,
"loss": 0.201,
"step": 1155
},
{
"epoch": 4.64,
"grad_norm": 2.342848968749728,
"learning_rate": 1.5708419435684463e-07,
"loss": 0.2045,
"step": 1160
},
{
"epoch": 4.66,
"grad_norm": 2.50611015184642,
"learning_rate": 1.4019499710726913e-07,
"loss": 0.2037,
"step": 1165
},
{
"epoch": 4.68,
"grad_norm": 1.9845962796916226,
"learning_rate": 1.2425322847218368e-07,
"loss": 0.1902,
"step": 1170
},
{
"epoch": 4.7,
"grad_norm": 2.066929322984826,
"learning_rate": 1.0926199633097156e-07,
"loss": 0.1956,
"step": 1175
},
{
"epoch": 4.72,
"grad_norm": 2.4138648473696853,
"learning_rate": 9.522422325404234e-08,
"loss": 0.1985,
"step": 1180
},
{
"epoch": 4.74,
"grad_norm": 2.911095557310394,
"learning_rate": 8.214264593307097e-08,
"loss": 0.1965,
"step": 1185
},
{
"epoch": 4.76,
"grad_norm": 2.030674439873152,
"learning_rate": 7.001981464747565e-08,
"loss": 0.1905,
"step": 1190
},
{
"epoch": 4.78,
"grad_norm": 2.5046761639631696,
"learning_rate": 5.8858092767236084e-08,
"loss": 0.2016,
"step": 1195
},
{
"epoch": 4.8,
"grad_norm": 2.6259164045048715,
"learning_rate": 4.865965629214819e-08,
"loss": 0.1991,
"step": 1200
},
{
"epoch": 4.82,
"grad_norm": 2.5096024708693725,
"learning_rate": 3.9426493427611177e-08,
"loss": 0.1987,
"step": 1205
},
{
"epoch": 4.84,
"grad_norm": 2.1146883842387783,
"learning_rate": 3.1160404197018155e-08,
"loss": 0.1915,
"step": 1210
},
{
"epoch": 4.86,
"grad_norm": 2.1798112190278136,
"learning_rate": 2.386300009084408e-08,
"loss": 0.2055,
"step": 1215
},
{
"epoch": 4.88,
"grad_norm": 2.1283754821360152,
"learning_rate": 1.753570375247815e-08,
"loss": 0.1941,
"step": 1220
},
{
"epoch": 4.9,
"grad_norm": 2.945625263690615,
"learning_rate": 1.2179748700879013e-08,
"loss": 0.2051,
"step": 1225
},
{
"epoch": 4.92,
"grad_norm": 2.3318033551646793,
"learning_rate": 7.796179090094891e-09,
"loss": 0.1929,
"step": 1230
},
{
"epoch": 4.9399999999999995,
"grad_norm": 2.7773560139669944,
"learning_rate": 4.385849505708084e-09,
"loss": 0.1958,
"step": 1235
},
{
"epoch": 4.96,
"grad_norm": 2.2518024677328685,
"learning_rate": 1.9494247982282386e-09,
"loss": 0.2032,
"step": 1240
},
{
"epoch": 4.98,
"grad_norm": 2.0203646274765132,
"learning_rate": 4.87379953478806e-10,
"loss": 0.2075,
"step": 1245
},
{
"epoch": 5.0,
"grad_norm": 2.9166287851342645,
"learning_rate": 0.0,
"loss": 0.1981,
"step": 1250
},
{
"epoch": 5.0,
"step": 1250,
"total_flos": 2442779688960.0,
"train_loss": 0.6922085983276367,
"train_runtime": 19109.4525,
"train_samples_per_second": 1.047,
"train_steps_per_second": 0.065
}
],
"logging_steps": 5,
"max_steps": 1250,
"num_input_tokens_seen": 0,
"num_train_epochs": 5,
"save_steps": 100,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 2442779688960.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}