1337d25c94
Model: xd2010/Qwen1.5-MOE-aux-free-sft-math7k-1e-3-gamma-part2-test Source: Original Platform
1764 lines
47 KiB
JSON
1764 lines
47 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.0,
|
|
"eval_steps": 500,
|
|
"global_step": 215,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"epoch": 0.004651162790697674,
|
|
"grad_norm": 2.2093663215637207,
|
|
"learning_rate": 0.0,
|
|
"loss": 0.2196,
|
|
"mean_token_accuracy": 0.9224496483802795,
|
|
"step": 1
|
|
},
|
|
{
|
|
"epoch": 0.009302325581395349,
|
|
"grad_norm": 2.072265625,
|
|
"learning_rate": 4.5454545454545457e-07,
|
|
"loss": 0.2166,
|
|
"mean_token_accuracy": 0.9229403138160706,
|
|
"step": 2
|
|
},
|
|
{
|
|
"epoch": 0.013953488372093023,
|
|
"grad_norm": 2.1760027408599854,
|
|
"learning_rate": 9.090909090909091e-07,
|
|
"loss": 0.2322,
|
|
"mean_token_accuracy": 0.9157230854034424,
|
|
"step": 3
|
|
},
|
|
{
|
|
"epoch": 0.018604651162790697,
|
|
"grad_norm": 2.0274431705474854,
|
|
"learning_rate": 1.3636363636363636e-06,
|
|
"loss": 0.2196,
|
|
"mean_token_accuracy": 0.9213500022888184,
|
|
"step": 4
|
|
},
|
|
{
|
|
"epoch": 0.023255813953488372,
|
|
"grad_norm": 2.3718724250793457,
|
|
"learning_rate": 1.8181818181818183e-06,
|
|
"loss": 0.2349,
|
|
"mean_token_accuracy": 0.9148977398872375,
|
|
"step": 5
|
|
},
|
|
{
|
|
"epoch": 0.027906976744186046,
|
|
"grad_norm": 2.291297197341919,
|
|
"learning_rate": 2.2727272727272728e-06,
|
|
"loss": 0.2199,
|
|
"mean_token_accuracy": 0.9249825477600098,
|
|
"step": 6
|
|
},
|
|
{
|
|
"epoch": 0.03255813953488372,
|
|
"grad_norm": 2.2355151176452637,
|
|
"learning_rate": 2.7272727272727272e-06,
|
|
"loss": 0.2184,
|
|
"mean_token_accuracy": 0.9212613105773926,
|
|
"step": 7
|
|
},
|
|
{
|
|
"epoch": 0.037209302325581395,
|
|
"grad_norm": 2.5275232791900635,
|
|
"learning_rate": 3.181818181818182e-06,
|
|
"loss": 0.2045,
|
|
"mean_token_accuracy": 0.927347719669342,
|
|
"step": 8
|
|
},
|
|
{
|
|
"epoch": 0.04186046511627907,
|
|
"grad_norm": 2.525670051574707,
|
|
"learning_rate": 3.6363636363636366e-06,
|
|
"loss": 0.1897,
|
|
"mean_token_accuracy": 0.9339665174484253,
|
|
"step": 9
|
|
},
|
|
{
|
|
"epoch": 0.046511627906976744,
|
|
"grad_norm": 2.0000622272491455,
|
|
"learning_rate": 4.0909090909090915e-06,
|
|
"loss": 0.1801,
|
|
"mean_token_accuracy": 0.9345327615737915,
|
|
"step": 10
|
|
},
|
|
{
|
|
"epoch": 0.05116279069767442,
|
|
"grad_norm": 1.8972795009613037,
|
|
"learning_rate": 4.5454545454545455e-06,
|
|
"loss": 0.1645,
|
|
"mean_token_accuracy": 0.9376271367073059,
|
|
"step": 11
|
|
},
|
|
{
|
|
"epoch": 0.05581395348837209,
|
|
"grad_norm": 1.9751707315444946,
|
|
"learning_rate": 5e-06,
|
|
"loss": 0.1706,
|
|
"mean_token_accuracy": 0.9384087324142456,
|
|
"step": 12
|
|
},
|
|
{
|
|
"epoch": 0.06046511627906977,
|
|
"grad_norm": 1.928964614868164,
|
|
"learning_rate": 5.4545454545454545e-06,
|
|
"loss": 0.1562,
|
|
"mean_token_accuracy": 0.9461650848388672,
|
|
"step": 13
|
|
},
|
|
{
|
|
"epoch": 0.06511627906976744,
|
|
"grad_norm": 1.9750374555587769,
|
|
"learning_rate": 5.90909090909091e-06,
|
|
"loss": 0.1223,
|
|
"mean_token_accuracy": 0.9516627192497253,
|
|
"step": 14
|
|
},
|
|
{
|
|
"epoch": 0.06976744186046512,
|
|
"grad_norm": 1.7492958307266235,
|
|
"learning_rate": 6.363636363636364e-06,
|
|
"loss": 0.0931,
|
|
"mean_token_accuracy": 0.9662150740623474,
|
|
"step": 15
|
|
},
|
|
{
|
|
"epoch": 0.07441860465116279,
|
|
"grad_norm": 2.532191753387451,
|
|
"learning_rate": 6.818181818181818e-06,
|
|
"loss": 0.1274,
|
|
"mean_token_accuracy": 0.9531450867652893,
|
|
"step": 16
|
|
},
|
|
{
|
|
"epoch": 0.07906976744186046,
|
|
"grad_norm": 2.4927215576171875,
|
|
"learning_rate": 7.272727272727273e-06,
|
|
"loss": 0.1033,
|
|
"mean_token_accuracy": 0.9629426598548889,
|
|
"step": 17
|
|
},
|
|
{
|
|
"epoch": 0.08372093023255814,
|
|
"grad_norm": 2.765667200088501,
|
|
"learning_rate": 7.727272727272727e-06,
|
|
"loss": 0.127,
|
|
"mean_token_accuracy": 0.9549099802970886,
|
|
"step": 18
|
|
},
|
|
{
|
|
"epoch": 0.08837209302325581,
|
|
"grad_norm": 3.610569715499878,
|
|
"learning_rate": 8.181818181818183e-06,
|
|
"loss": 0.1334,
|
|
"mean_token_accuracy": 0.9563636183738708,
|
|
"step": 19
|
|
},
|
|
{
|
|
"epoch": 0.09302325581395349,
|
|
"grad_norm": 2.513611316680908,
|
|
"learning_rate": 8.636363636363637e-06,
|
|
"loss": 0.1102,
|
|
"mean_token_accuracy": 0.9611197710037231,
|
|
"step": 20
|
|
},
|
|
{
|
|
"epoch": 0.09767441860465116,
|
|
"grad_norm": 2.8420629501342773,
|
|
"learning_rate": 9.090909090909091e-06,
|
|
"loss": 0.1243,
|
|
"mean_token_accuracy": 0.956632673740387,
|
|
"step": 21
|
|
},
|
|
{
|
|
"epoch": 0.10232558139534884,
|
|
"grad_norm": 2.757823944091797,
|
|
"learning_rate": 9.545454545454547e-06,
|
|
"loss": 0.1332,
|
|
"mean_token_accuracy": 0.9473860859870911,
|
|
"step": 22
|
|
},
|
|
{
|
|
"epoch": 0.10697674418604651,
|
|
"grad_norm": 2.2724320888519287,
|
|
"learning_rate": 1e-05,
|
|
"loss": 0.116,
|
|
"mean_token_accuracy": 0.958460807800293,
|
|
"step": 23
|
|
},
|
|
{
|
|
"epoch": 0.11162790697674418,
|
|
"grad_norm": 3.2801380157470703,
|
|
"learning_rate": 9.999403846557509e-06,
|
|
"loss": 0.1238,
|
|
"mean_token_accuracy": 0.9543752074241638,
|
|
"step": 24
|
|
},
|
|
{
|
|
"epoch": 0.11627906976744186,
|
|
"grad_norm": 2.1901867389678955,
|
|
"learning_rate": 9.99761554418511e-06,
|
|
"loss": 0.111,
|
|
"mean_token_accuracy": 0.9575592279434204,
|
|
"step": 25
|
|
},
|
|
{
|
|
"epoch": 0.12093023255813953,
|
|
"grad_norm": 2.26955246925354,
|
|
"learning_rate": 9.99463556670619e-06,
|
|
"loss": 0.1087,
|
|
"mean_token_accuracy": 0.9613473415374756,
|
|
"step": 26
|
|
},
|
|
{
|
|
"epoch": 0.12558139534883722,
|
|
"grad_norm": 2.3306360244750977,
|
|
"learning_rate": 9.990464703686895e-06,
|
|
"loss": 0.1308,
|
|
"mean_token_accuracy": 0.952462375164032,
|
|
"step": 27
|
|
},
|
|
{
|
|
"epoch": 0.13023255813953488,
|
|
"grad_norm": 2.3882932662963867,
|
|
"learning_rate": 9.985104060226937e-06,
|
|
"loss": 0.1229,
|
|
"mean_token_accuracy": 0.9562851786613464,
|
|
"step": 28
|
|
},
|
|
{
|
|
"epoch": 0.13488372093023257,
|
|
"grad_norm": 2.4214298725128174,
|
|
"learning_rate": 9.978555056666784e-06,
|
|
"loss": 0.1478,
|
|
"mean_token_accuracy": 0.9514402151107788,
|
|
"step": 29
|
|
},
|
|
{
|
|
"epoch": 0.13953488372093023,
|
|
"grad_norm": 2.251258373260498,
|
|
"learning_rate": 9.97081942821133e-06,
|
|
"loss": 0.1425,
|
|
"mean_token_accuracy": 0.9502054452896118,
|
|
"step": 30
|
|
},
|
|
{
|
|
"epoch": 0.14418604651162792,
|
|
"grad_norm": 2.2216179370880127,
|
|
"learning_rate": 9.961899224470146e-06,
|
|
"loss": 0.1441,
|
|
"mean_token_accuracy": 0.9497514367103577,
|
|
"step": 31
|
|
},
|
|
{
|
|
"epoch": 0.14883720930232558,
|
|
"grad_norm": 2.6528689861297607,
|
|
"learning_rate": 9.95179680891442e-06,
|
|
"loss": 0.117,
|
|
"mean_token_accuracy": 0.9592059850692749,
|
|
"step": 32
|
|
},
|
|
{
|
|
"epoch": 0.15348837209302327,
|
|
"grad_norm": 2.376858711242676,
|
|
"learning_rate": 9.940514858250736e-06,
|
|
"loss": 0.1199,
|
|
"mean_token_accuracy": 0.9578049778938293,
|
|
"step": 33
|
|
},
|
|
{
|
|
"epoch": 0.15813953488372093,
|
|
"grad_norm": 2.205568552017212,
|
|
"learning_rate": 9.928056361711854e-06,
|
|
"loss": 0.1197,
|
|
"mean_token_accuracy": 0.9574215412139893,
|
|
"step": 34
|
|
},
|
|
{
|
|
"epoch": 0.16279069767441862,
|
|
"grad_norm": 2.122694730758667,
|
|
"learning_rate": 9.914424620264714e-06,
|
|
"loss": 0.1337,
|
|
"mean_token_accuracy": 0.9519484639167786,
|
|
"step": 35
|
|
},
|
|
{
|
|
"epoch": 0.16744186046511628,
|
|
"grad_norm": 2.198246479034424,
|
|
"learning_rate": 9.899623245735798e-06,
|
|
"loss": 0.1196,
|
|
"mean_token_accuracy": 0.9565378427505493,
|
|
"step": 36
|
|
},
|
|
{
|
|
"epoch": 0.17209302325581396,
|
|
"grad_norm": 2.0603244304656982,
|
|
"learning_rate": 9.883656159854166e-06,
|
|
"loss": 0.1155,
|
|
"mean_token_accuracy": 0.9583718776702881,
|
|
"step": 37
|
|
},
|
|
{
|
|
"epoch": 0.17674418604651163,
|
|
"grad_norm": 2.1490249633789062,
|
|
"learning_rate": 9.866527593212355e-06,
|
|
"loss": 0.1149,
|
|
"mean_token_accuracy": 0.9578397870063782,
|
|
"step": 38
|
|
},
|
|
{
|
|
"epoch": 0.1813953488372093,
|
|
"grad_norm": 1.9495739936828613,
|
|
"learning_rate": 9.848242084145462e-06,
|
|
"loss": 0.1151,
|
|
"mean_token_accuracy": 0.9599069356918335,
|
|
"step": 39
|
|
},
|
|
{
|
|
"epoch": 0.18604651162790697,
|
|
"grad_norm": 2.2199368476867676,
|
|
"learning_rate": 9.82880447752868e-06,
|
|
"loss": 0.1258,
|
|
"mean_token_accuracy": 0.95333331823349,
|
|
"step": 40
|
|
},
|
|
{
|
|
"epoch": 0.19069767441860466,
|
|
"grad_norm": 2.1282341480255127,
|
|
"learning_rate": 9.808219923493606e-06,
|
|
"loss": 0.1175,
|
|
"mean_token_accuracy": 0.9576805233955383,
|
|
"step": 41
|
|
},
|
|
{
|
|
"epoch": 0.19534883720930232,
|
|
"grad_norm": 2.3917648792266846,
|
|
"learning_rate": 9.786493876063685e-06,
|
|
"loss": 0.1265,
|
|
"mean_token_accuracy": 0.9557332396507263,
|
|
"step": 42
|
|
},
|
|
{
|
|
"epoch": 0.2,
|
|
"grad_norm": 2.073394536972046,
|
|
"learning_rate": 9.763632091709125e-06,
|
|
"loss": 0.1077,
|
|
"mean_token_accuracy": 0.9598547220230103,
|
|
"step": 43
|
|
},
|
|
{
|
|
"epoch": 0.20465116279069767,
|
|
"grad_norm": 2.192857027053833,
|
|
"learning_rate": 9.739640627821678e-06,
|
|
"loss": 0.1296,
|
|
"mean_token_accuracy": 0.9574018716812134,
|
|
"step": 44
|
|
},
|
|
{
|
|
"epoch": 0.20930232558139536,
|
|
"grad_norm": 2.4736740589141846,
|
|
"learning_rate": 9.714525841109697e-06,
|
|
"loss": 0.1351,
|
|
"mean_token_accuracy": 0.9528483152389526,
|
|
"step": 45
|
|
},
|
|
{
|
|
"epoch": 0.21395348837209302,
|
|
"grad_norm": 2.248292922973633,
|
|
"learning_rate": 9.68829438591387e-06,
|
|
"loss": 0.1344,
|
|
"mean_token_accuracy": 0.9536008834838867,
|
|
"step": 46
|
|
},
|
|
{
|
|
"epoch": 0.2186046511627907,
|
|
"grad_norm": 2.214465856552124,
|
|
"learning_rate": 9.660953212444116e-06,
|
|
"loss": 0.1342,
|
|
"mean_token_accuracy": 0.9523285627365112,
|
|
"step": 47
|
|
},
|
|
{
|
|
"epoch": 0.22325581395348837,
|
|
"grad_norm": 2.3315677642822266,
|
|
"learning_rate": 9.632509564938073e-06,
|
|
"loss": 0.1365,
|
|
"mean_token_accuracy": 0.9518115520477295,
|
|
"step": 48
|
|
},
|
|
{
|
|
"epoch": 0.22790697674418606,
|
|
"grad_norm": 2.0210611820220947,
|
|
"learning_rate": 9.60297097974169e-06,
|
|
"loss": 0.1282,
|
|
"mean_token_accuracy": 0.9556302428245544,
|
|
"step": 49
|
|
},
|
|
{
|
|
"epoch": 0.23255813953488372,
|
|
"grad_norm": 2.0932199954986572,
|
|
"learning_rate": 9.572345283312407e-06,
|
|
"loss": 0.1125,
|
|
"mean_token_accuracy": 0.9600984454154968,
|
|
"step": 50
|
|
},
|
|
{
|
|
"epoch": 0.2372093023255814,
|
|
"grad_norm": 2.0512173175811768,
|
|
"learning_rate": 9.540640590145496e-06,
|
|
"loss": 0.1372,
|
|
"mean_token_accuracy": 0.9458490014076233,
|
|
"step": 51
|
|
},
|
|
{
|
|
"epoch": 0.24186046511627907,
|
|
"grad_norm": 2.433737277984619,
|
|
"learning_rate": 9.507865300624057e-06,
|
|
"loss": 0.1335,
|
|
"mean_token_accuracy": 0.9457087516784668,
|
|
"step": 52
|
|
},
|
|
{
|
|
"epoch": 0.24651162790697675,
|
|
"grad_norm": 2.981341600418091,
|
|
"learning_rate": 9.474028098793277e-06,
|
|
"loss": 0.1129,
|
|
"mean_token_accuracy": 0.9593979120254517,
|
|
"step": 53
|
|
},
|
|
{
|
|
"epoch": 0.25116279069767444,
|
|
"grad_norm": 2.2715203762054443,
|
|
"learning_rate": 9.439137950059539e-06,
|
|
"loss": 0.1333,
|
|
"mean_token_accuracy": 0.9539170265197754,
|
|
"step": 54
|
|
},
|
|
{
|
|
"epoch": 0.2558139534883721,
|
|
"grad_norm": 2.096383571624756,
|
|
"learning_rate": 9.403204098814965e-06,
|
|
"loss": 0.1443,
|
|
"mean_token_accuracy": 0.9474092721939087,
|
|
"step": 55
|
|
},
|
|
{
|
|
"epoch": 0.26046511627906976,
|
|
"grad_norm": 1.9958791732788086,
|
|
"learning_rate": 9.366236065988053e-06,
|
|
"loss": 0.1281,
|
|
"mean_token_accuracy": 0.9525547623634338,
|
|
"step": 56
|
|
},
|
|
{
|
|
"epoch": 0.2651162790697674,
|
|
"grad_norm": 2.0077316761016846,
|
|
"learning_rate": 9.32824364652104e-06,
|
|
"loss": 0.1208,
|
|
"mean_token_accuracy": 0.9563210606575012,
|
|
"step": 57
|
|
},
|
|
{
|
|
"epoch": 0.26976744186046514,
|
|
"grad_norm": 1.9917559623718262,
|
|
"learning_rate": 9.289236906774663e-06,
|
|
"loss": 0.1223,
|
|
"mean_token_accuracy": 0.9520987868309021,
|
|
"step": 58
|
|
},
|
|
{
|
|
"epoch": 0.2744186046511628,
|
|
"grad_norm": 2.441061019897461,
|
|
"learning_rate": 9.249226181861e-06,
|
|
"loss": 0.1384,
|
|
"mean_token_accuracy": 0.9503748416900635,
|
|
"step": 59
|
|
},
|
|
{
|
|
"epoch": 0.27906976744186046,
|
|
"grad_norm": 2.06553053855896,
|
|
"learning_rate": 9.208222072905113e-06,
|
|
"loss": 0.12,
|
|
"mean_token_accuracy": 0.9550289511680603,
|
|
"step": 60
|
|
},
|
|
{
|
|
"epoch": 0.2837209302325581,
|
|
"grad_norm": 2.4852378368377686,
|
|
"learning_rate": 9.166235444236209e-06,
|
|
"loss": 0.1497,
|
|
"mean_token_accuracy": 0.9450094699859619,
|
|
"step": 61
|
|
},
|
|
{
|
|
"epoch": 0.28837209302325584,
|
|
"grad_norm": 2.2705440521240234,
|
|
"learning_rate": 9.123277420509053e-06,
|
|
"loss": 0.1465,
|
|
"mean_token_accuracy": 0.9481350183486938,
|
|
"step": 62
|
|
},
|
|
{
|
|
"epoch": 0.2930232558139535,
|
|
"grad_norm": 2.179435968399048,
|
|
"learning_rate": 9.079359383756411e-06,
|
|
"loss": 0.1346,
|
|
"mean_token_accuracy": 0.9514462947845459,
|
|
"step": 63
|
|
},
|
|
{
|
|
"epoch": 0.29767441860465116,
|
|
"grad_norm": 2.21545147895813,
|
|
"learning_rate": 9.034492970373305e-06,
|
|
"loss": 0.128,
|
|
"mean_token_accuracy": 0.9546050429344177,
|
|
"step": 64
|
|
},
|
|
{
|
|
"epoch": 0.3023255813953488,
|
|
"grad_norm": 1.9919016361236572,
|
|
"learning_rate": 8.988690068033864e-06,
|
|
"loss": 0.1305,
|
|
"mean_token_accuracy": 0.954446017742157,
|
|
"step": 65
|
|
},
|
|
{
|
|
"epoch": 0.30697674418604654,
|
|
"grad_norm": 2.295849561691284,
|
|
"learning_rate": 8.941962812541604e-06,
|
|
"loss": 0.1289,
|
|
"mean_token_accuracy": 0.9543596506118774,
|
|
"step": 66
|
|
},
|
|
{
|
|
"epoch": 0.3116279069767442,
|
|
"grad_norm": 1.8786903619766235,
|
|
"learning_rate": 8.894323584613951e-06,
|
|
"loss": 0.1307,
|
|
"mean_token_accuracy": 0.9529724717140198,
|
|
"step": 67
|
|
},
|
|
{
|
|
"epoch": 0.31627906976744186,
|
|
"grad_norm": 1.99006986618042,
|
|
"learning_rate": 8.845785006601898e-06,
|
|
"loss": 0.1161,
|
|
"mean_token_accuracy": 0.9585253596305847,
|
|
"step": 68
|
|
},
|
|
{
|
|
"epoch": 0.3209302325581395,
|
|
"grad_norm": 2.211148738861084,
|
|
"learning_rate": 8.796359939145614e-06,
|
|
"loss": 0.1283,
|
|
"mean_token_accuracy": 0.9537134766578674,
|
|
"step": 69
|
|
},
|
|
{
|
|
"epoch": 0.32558139534883723,
|
|
"grad_norm": 2.460594415664673,
|
|
"learning_rate": 8.74606147776692e-06,
|
|
"loss": 0.1386,
|
|
"mean_token_accuracy": 0.9499496221542358,
|
|
"step": 70
|
|
},
|
|
{
|
|
"epoch": 0.3302325581395349,
|
|
"grad_norm": 2.2250773906707764,
|
|
"learning_rate": 8.694902949399555e-06,
|
|
"loss": 0.1368,
|
|
"mean_token_accuracy": 0.9525056481361389,
|
|
"step": 71
|
|
},
|
|
{
|
|
"epoch": 0.33488372093023255,
|
|
"grad_norm": 2.187652826309204,
|
|
"learning_rate": 8.642897908858096e-06,
|
|
"loss": 0.1299,
|
|
"mean_token_accuracy": 0.9521480798721313,
|
|
"step": 72
|
|
},
|
|
{
|
|
"epoch": 0.3395348837209302,
|
|
"grad_norm": 2.3103485107421875,
|
|
"learning_rate": 8.590060135246516e-06,
|
|
"loss": 0.1516,
|
|
"mean_token_accuracy": 0.944767415523529,
|
|
"step": 73
|
|
},
|
|
{
|
|
"epoch": 0.34418604651162793,
|
|
"grad_norm": 1.9994471073150635,
|
|
"learning_rate": 8.53640362830732e-06,
|
|
"loss": 0.1118,
|
|
"mean_token_accuracy": 0.9572561383247375,
|
|
"step": 74
|
|
},
|
|
{
|
|
"epoch": 0.3488372093023256,
|
|
"grad_norm": 2.2891528606414795,
|
|
"learning_rate": 8.481942604712209e-06,
|
|
"loss": 0.1361,
|
|
"mean_token_accuracy": 0.9537822604179382,
|
|
"step": 75
|
|
},
|
|
{
|
|
"epoch": 0.35348837209302325,
|
|
"grad_norm": 2.0732998847961426,
|
|
"learning_rate": 8.426691494295269e-06,
|
|
"loss": 0.1326,
|
|
"mean_token_accuracy": 0.9522348046302795,
|
|
"step": 76
|
|
},
|
|
{
|
|
"epoch": 0.3581395348837209,
|
|
"grad_norm": 2.020214557647705,
|
|
"learning_rate": 8.370664936229688e-06,
|
|
"loss": 0.1359,
|
|
"mean_token_accuracy": 0.9503569006919861,
|
|
"step": 77
|
|
},
|
|
{
|
|
"epoch": 0.3627906976744186,
|
|
"grad_norm": 2.1764564514160156,
|
|
"learning_rate": 8.313877775149009e-06,
|
|
"loss": 0.1245,
|
|
"mean_token_accuracy": 0.9583472013473511,
|
|
"step": 78
|
|
},
|
|
{
|
|
"epoch": 0.3674418604651163,
|
|
"grad_norm": 2.35109806060791,
|
|
"learning_rate": 8.256345057213925e-06,
|
|
"loss": 0.146,
|
|
"mean_token_accuracy": 0.9467026591300964,
|
|
"step": 79
|
|
},
|
|
{
|
|
"epoch": 0.37209302325581395,
|
|
"grad_norm": 2.0746405124664307,
|
|
"learning_rate": 8.198082026125707e-06,
|
|
"loss": 0.1263,
|
|
"mean_token_accuracy": 0.9523978233337402,
|
|
"step": 80
|
|
},
|
|
{
|
|
"epoch": 0.3767441860465116,
|
|
"grad_norm": 2.04236102104187,
|
|
"learning_rate": 8.139104119087265e-06,
|
|
"loss": 0.1255,
|
|
"mean_token_accuracy": 0.9556677937507629,
|
|
"step": 81
|
|
},
|
|
{
|
|
"epoch": 0.3813953488372093,
|
|
"grad_norm": 1.9404408931732178,
|
|
"learning_rate": 8.07942696271296e-06,
|
|
"loss": 0.1175,
|
|
"mean_token_accuracy": 0.9579678177833557,
|
|
"step": 82
|
|
},
|
|
{
|
|
"epoch": 0.386046511627907,
|
|
"grad_norm": 2.2659947872161865,
|
|
"learning_rate": 8.019066368888222e-06,
|
|
"loss": 0.1377,
|
|
"mean_token_accuracy": 0.9527695775032043,
|
|
"step": 83
|
|
},
|
|
{
|
|
"epoch": 0.39069767441860465,
|
|
"grad_norm": 2.161090850830078,
|
|
"learning_rate": 7.958038330580067e-06,
|
|
"loss": 0.1483,
|
|
"mean_token_accuracy": 0.9449127912521362,
|
|
"step": 84
|
|
},
|
|
{
|
|
"epoch": 0.3953488372093023,
|
|
"grad_norm": 1.9181567430496216,
|
|
"learning_rate": 7.89635901759967e-06,
|
|
"loss": 0.1379,
|
|
"mean_token_accuracy": 0.9500289559364319,
|
|
"step": 85
|
|
},
|
|
{
|
|
"epoch": 0.4,
|
|
"grad_norm": 2.3291268348693848,
|
|
"learning_rate": 7.834044772318033e-06,
|
|
"loss": 0.1677,
|
|
"mean_token_accuracy": 0.9397454261779785,
|
|
"step": 86
|
|
},
|
|
{
|
|
"epoch": 0.4046511627906977,
|
|
"grad_norm": 2.185105323791504,
|
|
"learning_rate": 7.77111210533597e-06,
|
|
"loss": 0.1325,
|
|
"mean_token_accuracy": 0.9541562795639038,
|
|
"step": 87
|
|
},
|
|
{
|
|
"epoch": 0.40930232558139534,
|
|
"grad_norm": 2.609286069869995,
|
|
"learning_rate": 7.707577691109519e-06,
|
|
"loss": 0.1806,
|
|
"mean_token_accuracy": 0.9326272010803223,
|
|
"step": 88
|
|
},
|
|
{
|
|
"epoch": 0.413953488372093,
|
|
"grad_norm": 2.013075828552246,
|
|
"learning_rate": 7.6434583635319e-06,
|
|
"loss": 0.139,
|
|
"mean_token_accuracy": 0.9526389241218567,
|
|
"step": 89
|
|
},
|
|
{
|
|
"epoch": 0.4186046511627907,
|
|
"grad_norm": 2.0093469619750977,
|
|
"learning_rate": 7.578771111473276e-06,
|
|
"loss": 0.1106,
|
|
"mean_token_accuracy": 0.9593625664710999,
|
|
"step": 90
|
|
},
|
|
{
|
|
"epoch": 0.4232558139534884,
|
|
"grad_norm": 1.9767247438430786,
|
|
"learning_rate": 7.513533074279427e-06,
|
|
"loss": 0.1157,
|
|
"mean_token_accuracy": 0.9587087035179138,
|
|
"step": 91
|
|
},
|
|
{
|
|
"epoch": 0.42790697674418604,
|
|
"grad_norm": 2.233933210372925,
|
|
"learning_rate": 7.4477615372305545e-06,
|
|
"loss": 0.138,
|
|
"mean_token_accuracy": 0.9505918622016907,
|
|
"step": 92
|
|
},
|
|
{
|
|
"epoch": 0.4325581395348837,
|
|
"grad_norm": 2.0568912029266357,
|
|
"learning_rate": 7.3814739269614265e-06,
|
|
"loss": 0.127,
|
|
"mean_token_accuracy": 0.9536969065666199,
|
|
"step": 93
|
|
},
|
|
{
|
|
"epoch": 0.4372093023255814,
|
|
"grad_norm": 2.0498812198638916,
|
|
"learning_rate": 7.314687806844067e-06,
|
|
"loss": 0.1423,
|
|
"mean_token_accuracy": 0.9466473460197449,
|
|
"step": 94
|
|
},
|
|
{
|
|
"epoch": 0.4418604651162791,
|
|
"grad_norm": 2.2531538009643555,
|
|
"learning_rate": 7.247420872334221e-06,
|
|
"loss": 0.1349,
|
|
"mean_token_accuracy": 0.9516788125038147,
|
|
"step": 95
|
|
},
|
|
{
|
|
"epoch": 0.44651162790697674,
|
|
"grad_norm": 1.951317310333252,
|
|
"learning_rate": 7.179690946282808e-06,
|
|
"loss": 0.1162,
|
|
"mean_token_accuracy": 0.9578741788864136,
|
|
"step": 96
|
|
},
|
|
{
|
|
"epoch": 0.4511627906976744,
|
|
"grad_norm": 2.0403988361358643,
|
|
"learning_rate": 7.111515974213639e-06,
|
|
"loss": 0.1175,
|
|
"mean_token_accuracy": 0.9583046436309814,
|
|
"step": 97
|
|
},
|
|
{
|
|
"epoch": 0.4558139534883721,
|
|
"grad_norm": 2.085268497467041,
|
|
"learning_rate": 7.042914019568621e-06,
|
|
"loss": 0.1319,
|
|
"mean_token_accuracy": 0.9513552188873291,
|
|
"step": 98
|
|
},
|
|
{
|
|
"epoch": 0.4604651162790698,
|
|
"grad_norm": 1.830336093902588,
|
|
"learning_rate": 6.973903258921719e-06,
|
|
"loss": 0.118,
|
|
"mean_token_accuracy": 0.9577862620353699,
|
|
"step": 99
|
|
},
|
|
{
|
|
"epoch": 0.46511627906976744,
|
|
"grad_norm": 2.2048964500427246,
|
|
"learning_rate": 6.904501977162949e-06,
|
|
"loss": 0.146,
|
|
"mean_token_accuracy": 0.9484704732894897,
|
|
"step": 100
|
|
},
|
|
{
|
|
"epoch": 0.4697674418604651,
|
|
"grad_norm": 2.0180842876434326,
|
|
"learning_rate": 6.834728562653659e-06,
|
|
"loss": 0.1406,
|
|
"mean_token_accuracy": 0.9502673149108887,
|
|
"step": 101
|
|
},
|
|
{
|
|
"epoch": 0.4744186046511628,
|
|
"grad_norm": 2.2566816806793213,
|
|
"learning_rate": 6.764601502354403e-06,
|
|
"loss": 0.14,
|
|
"mean_token_accuracy": 0.9521723985671997,
|
|
"step": 102
|
|
},
|
|
{
|
|
"epoch": 0.4790697674418605,
|
|
"grad_norm": 2.162036895751953,
|
|
"learning_rate": 6.6941393769266995e-06,
|
|
"loss": 0.1509,
|
|
"mean_token_accuracy": 0.9466860890388489,
|
|
"step": 103
|
|
},
|
|
{
|
|
"epoch": 0.48372093023255813,
|
|
"grad_norm": 2.131171703338623,
|
|
"learning_rate": 6.6233608558099405e-06,
|
|
"loss": 0.129,
|
|
"mean_token_accuracy": 0.9520898461341858,
|
|
"step": 104
|
|
},
|
|
{
|
|
"epoch": 0.4883720930232558,
|
|
"grad_norm": 2.283207654953003,
|
|
"learning_rate": 6.552284692274803e-06,
|
|
"loss": 0.1592,
|
|
"mean_token_accuracy": 0.9394658803939819,
|
|
"step": 105
|
|
},
|
|
{
|
|
"epoch": 0.4930232558139535,
|
|
"grad_norm": 2.2977681159973145,
|
|
"learning_rate": 6.48092971845443e-06,
|
|
"loss": 0.1554,
|
|
"mean_token_accuracy": 0.9446787238121033,
|
|
"step": 106
|
|
},
|
|
{
|
|
"epoch": 0.49767441860465117,
|
|
"grad_norm": 2.474445104598999,
|
|
"learning_rate": 6.409314840354724e-06,
|
|
"loss": 0.1764,
|
|
"mean_token_accuracy": 0.9388563632965088,
|
|
"step": 107
|
|
},
|
|
{
|
|
"epoch": 0.5023255813953489,
|
|
"grad_norm": 2.3474838733673096,
|
|
"learning_rate": 6.337459032845068e-06,
|
|
"loss": 0.1786,
|
|
"mean_token_accuracy": 0.9278457164764404,
|
|
"step": 108
|
|
},
|
|
{
|
|
"epoch": 0.5069767441860465,
|
|
"grad_norm": 2.1931817531585693,
|
|
"learning_rate": 6.2653813346308e-06,
|
|
"loss": 0.1511,
|
|
"mean_token_accuracy": 0.9489392042160034,
|
|
"step": 109
|
|
},
|
|
{
|
|
"epoch": 0.5116279069767442,
|
|
"grad_norm": 2.1758244037628174,
|
|
"learning_rate": 6.193100843208772e-06,
|
|
"loss": 0.1479,
|
|
"mean_token_accuracy": 0.9441016912460327,
|
|
"step": 110
|
|
},
|
|
{
|
|
"epoch": 0.5162790697674419,
|
|
"grad_norm": 2.205124616622925,
|
|
"learning_rate": 6.120636709807334e-06,
|
|
"loss": 0.1602,
|
|
"mean_token_accuracy": 0.9404239058494568,
|
|
"step": 111
|
|
},
|
|
{
|
|
"epoch": 0.5209302325581395,
|
|
"grad_norm": 2.4089300632476807,
|
|
"learning_rate": 6.048008134312078e-06,
|
|
"loss": 0.1537,
|
|
"mean_token_accuracy": 0.9437205195426941,
|
|
"step": 112
|
|
},
|
|
{
|
|
"epoch": 0.5255813953488372,
|
|
"grad_norm": 2.23669695854187,
|
|
"learning_rate": 5.975234360178698e-06,
|
|
"loss": 0.1876,
|
|
"mean_token_accuracy": 0.9296743273735046,
|
|
"step": 113
|
|
},
|
|
{
|
|
"epoch": 0.5302325581395348,
|
|
"grad_norm": 2.0364840030670166,
|
|
"learning_rate": 5.902334669334287e-06,
|
|
"loss": 0.1575,
|
|
"mean_token_accuracy": 0.9448131322860718,
|
|
"step": 114
|
|
},
|
|
{
|
|
"epoch": 0.5348837209302325,
|
|
"grad_norm": 2.1594290733337402,
|
|
"learning_rate": 5.829328377068476e-06,
|
|
"loss": 0.1631,
|
|
"mean_token_accuracy": 0.9371219277381897,
|
|
"step": 115
|
|
},
|
|
{
|
|
"epoch": 0.5395348837209303,
|
|
"grad_norm": 2.3116836547851562,
|
|
"learning_rate": 5.756234826915686e-06,
|
|
"loss": 0.178,
|
|
"mean_token_accuracy": 0.9351682066917419,
|
|
"step": 116
|
|
},
|
|
{
|
|
"epoch": 0.5441860465116279,
|
|
"grad_norm": 2.4108450412750244,
|
|
"learning_rate": 5.683073385529938e-06,
|
|
"loss": 0.1671,
|
|
"mean_token_accuracy": 0.9367252588272095,
|
|
"step": 117
|
|
},
|
|
{
|
|
"epoch": 0.5488372093023256,
|
|
"grad_norm": 2.075869083404541,
|
|
"learning_rate": 5.60986343755352e-06,
|
|
"loss": 0.1472,
|
|
"mean_token_accuracy": 0.9467573165893555,
|
|
"step": 118
|
|
},
|
|
{
|
|
"epoch": 0.5534883720930233,
|
|
"grad_norm": 2.02726674079895,
|
|
"learning_rate": 5.536624380480878e-06,
|
|
"loss": 0.1385,
|
|
"mean_token_accuracy": 0.9516604542732239,
|
|
"step": 119
|
|
},
|
|
{
|
|
"epoch": 0.5581395348837209,
|
|
"grad_norm": 2.022648334503174,
|
|
"learning_rate": 5.4633756195191235e-06,
|
|
"loss": 0.1455,
|
|
"mean_token_accuracy": 0.9479925632476807,
|
|
"step": 120
|
|
},
|
|
{
|
|
"epoch": 0.5627906976744186,
|
|
"grad_norm": 1.7579838037490845,
|
|
"learning_rate": 5.390136562446482e-06,
|
|
"loss": 0.13,
|
|
"mean_token_accuracy": 0.9530401229858398,
|
|
"step": 121
|
|
},
|
|
{
|
|
"epoch": 0.5674418604651162,
|
|
"grad_norm": 2.16518497467041,
|
|
"learning_rate": 5.316926614470063e-06,
|
|
"loss": 0.166,
|
|
"mean_token_accuracy": 0.9389764070510864,
|
|
"step": 122
|
|
},
|
|
{
|
|
"epoch": 0.5720930232558139,
|
|
"grad_norm": 2.500831365585327,
|
|
"learning_rate": 5.2437651730843165e-06,
|
|
"loss": 0.2113,
|
|
"mean_token_accuracy": 0.9219081997871399,
|
|
"step": 123
|
|
},
|
|
{
|
|
"epoch": 0.5767441860465117,
|
|
"grad_norm": 2.5281460285186768,
|
|
"learning_rate": 5.170671622931527e-06,
|
|
"loss": 0.2038,
|
|
"mean_token_accuracy": 0.9247100949287415,
|
|
"step": 124
|
|
},
|
|
{
|
|
"epoch": 0.5813953488372093,
|
|
"grad_norm": 2.025871515274048,
|
|
"learning_rate": 5.097665330665714e-06,
|
|
"loss": 0.125,
|
|
"mean_token_accuracy": 0.9528625011444092,
|
|
"step": 125
|
|
},
|
|
{
|
|
"epoch": 0.586046511627907,
|
|
"grad_norm": 2.469729423522949,
|
|
"learning_rate": 5.024765639821305e-06,
|
|
"loss": 0.1749,
|
|
"mean_token_accuracy": 0.9346743226051331,
|
|
"step": 126
|
|
},
|
|
{
|
|
"epoch": 0.5906976744186047,
|
|
"grad_norm": 1.9095555543899536,
|
|
"learning_rate": 4.951991865687923e-06,
|
|
"loss": 0.1378,
|
|
"mean_token_accuracy": 0.95150226354599,
|
|
"step": 127
|
|
},
|
|
{
|
|
"epoch": 0.5953488372093023,
|
|
"grad_norm": 2.488006591796875,
|
|
"learning_rate": 4.879363290192667e-06,
|
|
"loss": 0.1836,
|
|
"mean_token_accuracy": 0.9379990100860596,
|
|
"step": 128
|
|
},
|
|
{
|
|
"epoch": 0.6,
|
|
"grad_norm": 2.5464515686035156,
|
|
"learning_rate": 4.806899156791231e-06,
|
|
"loss": 0.2087,
|
|
"mean_token_accuracy": 0.9257314205169678,
|
|
"step": 129
|
|
},
|
|
{
|
|
"epoch": 0.6046511627906976,
|
|
"grad_norm": 2.5360970497131348,
|
|
"learning_rate": 4.734618665369202e-06,
|
|
"loss": 0.1761,
|
|
"mean_token_accuracy": 0.933690071105957,
|
|
"step": 130
|
|
},
|
|
{
|
|
"epoch": 0.6093023255813953,
|
|
"grad_norm": 2.1590800285339355,
|
|
"learning_rate": 4.662540967154934e-06,
|
|
"loss": 0.1575,
|
|
"mean_token_accuracy": 0.9439483284950256,
|
|
"step": 131
|
|
},
|
|
{
|
|
"epoch": 0.6139534883720931,
|
|
"grad_norm": 2.1707804203033447,
|
|
"learning_rate": 4.5906851596452765e-06,
|
|
"loss": 0.1604,
|
|
"mean_token_accuracy": 0.9432588815689087,
|
|
"step": 132
|
|
},
|
|
{
|
|
"epoch": 0.6186046511627907,
|
|
"grad_norm": 2.399054527282715,
|
|
"learning_rate": 4.519070281545571e-06,
|
|
"loss": 0.1858,
|
|
"mean_token_accuracy": 0.93243008852005,
|
|
"step": 133
|
|
},
|
|
{
|
|
"epoch": 0.6232558139534884,
|
|
"grad_norm": 2.4510555267333984,
|
|
"learning_rate": 4.447715307725197e-06,
|
|
"loss": 0.1835,
|
|
"mean_token_accuracy": 0.9314557313919067,
|
|
"step": 134
|
|
},
|
|
{
|
|
"epoch": 0.627906976744186,
|
|
"grad_norm": 2.3069207668304443,
|
|
"learning_rate": 4.376639144190061e-06,
|
|
"loss": 0.1989,
|
|
"mean_token_accuracy": 0.9328960180282593,
|
|
"step": 135
|
|
},
|
|
{
|
|
"epoch": 0.6325581395348837,
|
|
"grad_norm": 2.2998270988464355,
|
|
"learning_rate": 4.305860623073304e-06,
|
|
"loss": 0.1881,
|
|
"mean_token_accuracy": 0.9304514527320862,
|
|
"step": 136
|
|
},
|
|
{
|
|
"epoch": 0.6372093023255814,
|
|
"grad_norm": 2.1398470401763916,
|
|
"learning_rate": 4.2353984976456e-06,
|
|
"loss": 0.1644,
|
|
"mean_token_accuracy": 0.9427598118782043,
|
|
"step": 137
|
|
},
|
|
{
|
|
"epoch": 0.641860465116279,
|
|
"grad_norm": 2.2165541648864746,
|
|
"learning_rate": 4.1652714373463435e-06,
|
|
"loss": 0.1745,
|
|
"mean_token_accuracy": 0.9365646243095398,
|
|
"step": 138
|
|
},
|
|
{
|
|
"epoch": 0.6465116279069767,
|
|
"grad_norm": 2.1547625064849854,
|
|
"learning_rate": 4.095498022837051e-06,
|
|
"loss": 0.1745,
|
|
"mean_token_accuracy": 0.9377639293670654,
|
|
"step": 139
|
|
},
|
|
{
|
|
"epoch": 0.6511627906976745,
|
|
"grad_norm": 2.40480899810791,
|
|
"learning_rate": 4.026096741078281e-06,
|
|
"loss": 0.2065,
|
|
"mean_token_accuracy": 0.9249402284622192,
|
|
"step": 140
|
|
},
|
|
{
|
|
"epoch": 0.6558139534883721,
|
|
"grad_norm": 2.3467397689819336,
|
|
"learning_rate": 3.957085980431382e-06,
|
|
"loss": 0.2017,
|
|
"mean_token_accuracy": 0.9295312762260437,
|
|
"step": 141
|
|
},
|
|
{
|
|
"epoch": 0.6604651162790698,
|
|
"grad_norm": 2.40217924118042,
|
|
"learning_rate": 3.888484025786364e-06,
|
|
"loss": 0.1892,
|
|
"mean_token_accuracy": 0.93282550573349,
|
|
"step": 142
|
|
},
|
|
{
|
|
"epoch": 0.6651162790697674,
|
|
"grad_norm": 2.2477176189422607,
|
|
"learning_rate": 3.820309053717195e-06,
|
|
"loss": 0.1714,
|
|
"mean_token_accuracy": 0.9409193992614746,
|
|
"step": 143
|
|
},
|
|
{
|
|
"epoch": 0.6697674418604651,
|
|
"grad_norm": 2.544952630996704,
|
|
"learning_rate": 3.75257912766578e-06,
|
|
"loss": 0.2318,
|
|
"mean_token_accuracy": 0.9194183349609375,
|
|
"step": 144
|
|
},
|
|
{
|
|
"epoch": 0.6744186046511628,
|
|
"grad_norm": 2.7415833473205566,
|
|
"learning_rate": 3.6853121931559334e-06,
|
|
"loss": 0.215,
|
|
"mean_token_accuracy": 0.9232286810874939,
|
|
"step": 145
|
|
},
|
|
{
|
|
"epoch": 0.6790697674418604,
|
|
"grad_norm": 2.43613600730896,
|
|
"learning_rate": 3.618526073038574e-06,
|
|
"loss": 0.2455,
|
|
"mean_token_accuracy": 0.9059690237045288,
|
|
"step": 146
|
|
},
|
|
{
|
|
"epoch": 0.6837209302325581,
|
|
"grad_norm": 2.250971794128418,
|
|
"learning_rate": 3.552238462769446e-06,
|
|
"loss": 0.1811,
|
|
"mean_token_accuracy": 0.9321259260177612,
|
|
"step": 147
|
|
},
|
|
{
|
|
"epoch": 0.6883720930232559,
|
|
"grad_norm": 2.219188690185547,
|
|
"learning_rate": 3.4864669257205745e-06,
|
|
"loss": 0.1832,
|
|
"mean_token_accuracy": 0.9357648491859436,
|
|
"step": 148
|
|
},
|
|
{
|
|
"epoch": 0.6930232558139535,
|
|
"grad_norm": 2.256803512573242,
|
|
"learning_rate": 3.4212288885267246e-06,
|
|
"loss": 0.2102,
|
|
"mean_token_accuracy": 0.9252863526344299,
|
|
"step": 149
|
|
},
|
|
{
|
|
"epoch": 0.6976744186046512,
|
|
"grad_norm": 2.392669916152954,
|
|
"learning_rate": 3.3565416364681016e-06,
|
|
"loss": 0.1942,
|
|
"mean_token_accuracy": 0.931359589099884,
|
|
"step": 150
|
|
},
|
|
{
|
|
"epoch": 0.7023255813953488,
|
|
"grad_norm": 2.0853817462921143,
|
|
"learning_rate": 3.2924223088904816e-06,
|
|
"loss": 0.1826,
|
|
"mean_token_accuracy": 0.9341450929641724,
|
|
"step": 151
|
|
},
|
|
{
|
|
"epoch": 0.7069767441860465,
|
|
"grad_norm": 2.7918574810028076,
|
|
"learning_rate": 3.228887894664029e-06,
|
|
"loss": 0.2417,
|
|
"mean_token_accuracy": 0.9140521287918091,
|
|
"step": 152
|
|
},
|
|
{
|
|
"epoch": 0.7116279069767442,
|
|
"grad_norm": 2.603607177734375,
|
|
"learning_rate": 3.1659552276819693e-06,
|
|
"loss": 0.2286,
|
|
"mean_token_accuracy": 0.9171662330627441,
|
|
"step": 153
|
|
},
|
|
{
|
|
"epoch": 0.7162790697674418,
|
|
"grad_norm": 3.1204798221588135,
|
|
"learning_rate": 3.1036409824003324e-06,
|
|
"loss": 0.2008,
|
|
"mean_token_accuracy": 0.9280551075935364,
|
|
"step": 154
|
|
},
|
|
{
|
|
"epoch": 0.7209302325581395,
|
|
"grad_norm": 2.186347723007202,
|
|
"learning_rate": 3.0419616694199327e-06,
|
|
"loss": 0.1883,
|
|
"mean_token_accuracy": 0.9328477382659912,
|
|
"step": 155
|
|
},
|
|
{
|
|
"epoch": 0.7255813953488373,
|
|
"grad_norm": 2.160309076309204,
|
|
"learning_rate": 2.98093363111178e-06,
|
|
"loss": 0.1594,
|
|
"mean_token_accuracy": 0.9413360953330994,
|
|
"step": 156
|
|
},
|
|
{
|
|
"epoch": 0.7302325581395349,
|
|
"grad_norm": 2.4859275817871094,
|
|
"learning_rate": 2.92057303728704e-06,
|
|
"loss": 0.2126,
|
|
"mean_token_accuracy": 0.9240214824676514,
|
|
"step": 157
|
|
},
|
|
{
|
|
"epoch": 0.7348837209302326,
|
|
"grad_norm": 2.2752139568328857,
|
|
"learning_rate": 2.860895880912735e-06,
|
|
"loss": 0.1996,
|
|
"mean_token_accuracy": 0.9289940595626831,
|
|
"step": 158
|
|
},
|
|
{
|
|
"epoch": 0.7395348837209302,
|
|
"grad_norm": 2.431081771850586,
|
|
"learning_rate": 2.801917973874294e-06,
|
|
"loss": 0.2276,
|
|
"mean_token_accuracy": 0.9175781607627869,
|
|
"step": 159
|
|
},
|
|
{
|
|
"epoch": 0.7441860465116279,
|
|
"grad_norm": 2.5055429935455322,
|
|
"learning_rate": 2.7436549427860766e-06,
|
|
"loss": 0.2413,
|
|
"mean_token_accuracy": 0.914275586605072,
|
|
"step": 160
|
|
},
|
|
{
|
|
"epoch": 0.7488372093023256,
|
|
"grad_norm": 2.182417392730713,
|
|
"learning_rate": 2.6861222248509926e-06,
|
|
"loss": 0.1774,
|
|
"mean_token_accuracy": 0.9350738525390625,
|
|
"step": 161
|
|
},
|
|
{
|
|
"epoch": 0.7534883720930232,
|
|
"grad_norm": 2.3123719692230225,
|
|
"learning_rate": 2.6293350637703123e-06,
|
|
"loss": 0.2111,
|
|
"mean_token_accuracy": 0.9277293682098389,
|
|
"step": 162
|
|
},
|
|
{
|
|
"epoch": 0.7581395348837209,
|
|
"grad_norm": 2.4065475463867188,
|
|
"learning_rate": 2.5733085057047325e-06,
|
|
"loss": 0.2135,
|
|
"mean_token_accuracy": 0.9246734380722046,
|
|
"step": 163
|
|
},
|
|
{
|
|
"epoch": 0.7627906976744186,
|
|
"grad_norm": 2.3099215030670166,
|
|
"learning_rate": 2.518057395287792e-06,
|
|
"loss": 0.2293,
|
|
"mean_token_accuracy": 0.9176149964332581,
|
|
"step": 164
|
|
},
|
|
{
|
|
"epoch": 0.7674418604651163,
|
|
"grad_norm": 2.545142889022827,
|
|
"learning_rate": 2.463596371692681e-06,
|
|
"loss": 0.2269,
|
|
"mean_token_accuracy": 0.9170432090759277,
|
|
"step": 165
|
|
},
|
|
{
|
|
"epoch": 0.772093023255814,
|
|
"grad_norm": 2.567596912384033,
|
|
"learning_rate": 2.409939864753487e-06,
|
|
"loss": 0.2417,
|
|
"mean_token_accuracy": 0.9141337275505066,
|
|
"step": 166
|
|
},
|
|
{
|
|
"epoch": 0.7767441860465116,
|
|
"grad_norm": 2.2670516967773438,
|
|
"learning_rate": 2.3571020911419067e-06,
|
|
"loss": 0.2282,
|
|
"mean_token_accuracy": 0.9241269826889038,
|
|
"step": 167
|
|
},
|
|
{
|
|
"epoch": 0.7813953488372093,
|
|
"grad_norm": 2.5289041996002197,
|
|
"learning_rate": 2.3050970506004463e-06,
|
|
"loss": 0.2492,
|
|
"mean_token_accuracy": 0.9122055768966675,
|
|
"step": 168
|
|
},
|
|
{
|
|
"epoch": 0.786046511627907,
|
|
"grad_norm": 2.4541704654693604,
|
|
"learning_rate": 2.2539385222330797e-06,
|
|
"loss": 0.1962,
|
|
"mean_token_accuracy": 0.929431676864624,
|
|
"step": 169
|
|
},
|
|
{
|
|
"epoch": 0.7906976744186046,
|
|
"grad_norm": 2.301494836807251,
|
|
"learning_rate": 2.203640060854387e-06,
|
|
"loss": 0.2171,
|
|
"mean_token_accuracy": 0.9187138676643372,
|
|
"step": 170
|
|
},
|
|
{
|
|
"epoch": 0.7953488372093023,
|
|
"grad_norm": 3.4231297969818115,
|
|
"learning_rate": 2.1542149933981014e-06,
|
|
"loss": 0.2418,
|
|
"mean_token_accuracy": 0.912670373916626,
|
|
"step": 171
|
|
},
|
|
{
|
|
"epoch": 0.8,
|
|
"grad_norm": 2.385003089904785,
|
|
"learning_rate": 2.10567641538605e-06,
|
|
"loss": 0.2195,
|
|
"mean_token_accuracy": 0.9225817918777466,
|
|
"step": 172
|
|
},
|
|
{
|
|
"epoch": 0.8046511627906977,
|
|
"grad_norm": 2.3375015258789062,
|
|
"learning_rate": 2.058037187458398e-06,
|
|
"loss": 0.2348,
|
|
"mean_token_accuracy": 0.9222442507743835,
|
|
"step": 173
|
|
},
|
|
{
|
|
"epoch": 0.8093023255813954,
|
|
"grad_norm": 2.641136884689331,
|
|
"learning_rate": 2.011309931966136e-06,
|
|
"loss": 0.2418,
|
|
"mean_token_accuracy": 0.9168000221252441,
|
|
"step": 174
|
|
},
|
|
{
|
|
"epoch": 0.813953488372093,
|
|
"grad_norm": 2.500105619430542,
|
|
"learning_rate": 1.965507029626695e-06,
|
|
"loss": 0.2529,
|
|
"mean_token_accuracy": 0.8996127247810364,
|
|
"step": 175
|
|
},
|
|
{
|
|
"epoch": 0.8186046511627907,
|
|
"grad_norm": 2.368457555770874,
|
|
"learning_rate": 1.920640616243589e-06,
|
|
"loss": 0.2154,
|
|
"mean_token_accuracy": 0.9224004149436951,
|
|
"step": 176
|
|
},
|
|
{
|
|
"epoch": 0.8232558139534883,
|
|
"grad_norm": 2.742520809173584,
|
|
"learning_rate": 1.8767225794909484e-06,
|
|
"loss": 0.2791,
|
|
"mean_token_accuracy": 0.9038854837417603,
|
|
"step": 177
|
|
},
|
|
{
|
|
"epoch": 0.827906976744186,
|
|
"grad_norm": 2.3562746047973633,
|
|
"learning_rate": 1.8337645557637929e-06,
|
|
"loss": 0.233,
|
|
"mean_token_accuracy": 0.9172508120536804,
|
|
"step": 178
|
|
},
|
|
{
|
|
"epoch": 0.8325581395348837,
|
|
"grad_norm": 2.310187816619873,
|
|
"learning_rate": 1.7917779270948887e-06,
|
|
"loss": 0.2517,
|
|
"mean_token_accuracy": 0.9111905694007874,
|
|
"step": 179
|
|
},
|
|
{
|
|
"epoch": 0.8372093023255814,
|
|
"grad_norm": 2.510735511779785,
|
|
"learning_rate": 1.7507738181390027e-06,
|
|
"loss": 0.2461,
|
|
"mean_token_accuracy": 0.9119247794151306,
|
|
"step": 180
|
|
},
|
|
{
|
|
"epoch": 0.8418604651162791,
|
|
"grad_norm": 2.414520502090454,
|
|
"learning_rate": 1.7107630932253383e-06,
|
|
"loss": 0.2419,
|
|
"mean_token_accuracy": 0.9190771579742432,
|
|
"step": 181
|
|
},
|
|
{
|
|
"epoch": 0.8465116279069768,
|
|
"grad_norm": 2.4094643592834473,
|
|
"learning_rate": 1.6717563534789594e-06,
|
|
"loss": 0.2469,
|
|
"mean_token_accuracy": 0.9117172360420227,
|
|
"step": 182
|
|
},
|
|
{
|
|
"epoch": 0.8511627906976744,
|
|
"grad_norm": 2.364867925643921,
|
|
"learning_rate": 1.6337639340119476e-06,
|
|
"loss": 0.2644,
|
|
"mean_token_accuracy": 0.9048944711685181,
|
|
"step": 183
|
|
},
|
|
{
|
|
"epoch": 0.8558139534883721,
|
|
"grad_norm": 2.6265275478363037,
|
|
"learning_rate": 1.596795901185037e-06,
|
|
"loss": 0.2633,
|
|
"mean_token_accuracy": 0.9058024287223816,
|
|
"step": 184
|
|
},
|
|
{
|
|
"epoch": 0.8604651162790697,
|
|
"grad_norm": 2.300539255142212,
|
|
"learning_rate": 1.5608620499404628e-06,
|
|
"loss": 0.2402,
|
|
"mean_token_accuracy": 0.9132765531539917,
|
|
"step": 185
|
|
},
|
|
{
|
|
"epoch": 0.8651162790697674,
|
|
"grad_norm": 2.4773666858673096,
|
|
"learning_rate": 1.5259719012067249e-06,
|
|
"loss": 0.2646,
|
|
"mean_token_accuracy": 0.9078861474990845,
|
|
"step": 186
|
|
},
|
|
{
|
|
"epoch": 0.8697674418604651,
|
|
"grad_norm": 2.367901563644409,
|
|
"learning_rate": 1.4921346993759453e-06,
|
|
"loss": 0.2409,
|
|
"mean_token_accuracy": 0.9140973091125488,
|
|
"step": 187
|
|
},
|
|
{
|
|
"epoch": 0.8744186046511628,
|
|
"grad_norm": 2.2784316539764404,
|
|
"learning_rate": 1.459359409854505e-06,
|
|
"loss": 0.2481,
|
|
"mean_token_accuracy": 0.9134429693222046,
|
|
"step": 188
|
|
},
|
|
{
|
|
"epoch": 0.8790697674418605,
|
|
"grad_norm": 2.370464324951172,
|
|
"learning_rate": 1.4276547166875946e-06,
|
|
"loss": 0.2355,
|
|
"mean_token_accuracy": 0.9189856648445129,
|
|
"step": 189
|
|
},
|
|
{
|
|
"epoch": 0.8837209302325582,
|
|
"grad_norm": 2.743762969970703,
|
|
"learning_rate": 1.397029020258313e-06,
|
|
"loss": 0.2568,
|
|
"mean_token_accuracy": 0.90625,
|
|
"step": 190
|
|
},
|
|
{
|
|
"epoch": 0.8883720930232558,
|
|
"grad_norm": 2.4993607997894287,
|
|
"learning_rate": 1.367490435061928e-06,
|
|
"loss": 0.2989,
|
|
"mean_token_accuracy": 0.8918637633323669,
|
|
"step": 191
|
|
},
|
|
{
|
|
"epoch": 0.8930232558139535,
|
|
"grad_norm": 3.4072718620300293,
|
|
"learning_rate": 1.3390467875558855e-06,
|
|
"loss": 0.2207,
|
|
"mean_token_accuracy": 0.9221254587173462,
|
|
"step": 192
|
|
},
|
|
{
|
|
"epoch": 0.8976744186046511,
|
|
"grad_norm": 2.3936197757720947,
|
|
"learning_rate": 1.3117056140861317e-06,
|
|
"loss": 0.2658,
|
|
"mean_token_accuracy": 0.9089629054069519,
|
|
"step": 193
|
|
},
|
|
{
|
|
"epoch": 0.9023255813953488,
|
|
"grad_norm": 2.5090274810791016,
|
|
"learning_rate": 1.285474158890304e-06,
|
|
"loss": 0.2309,
|
|
"mean_token_accuracy": 0.9217953085899353,
|
|
"step": 194
|
|
},
|
|
{
|
|
"epoch": 0.9069767441860465,
|
|
"grad_norm": 2.9065701961517334,
|
|
"learning_rate": 1.2603593721783219e-06,
|
|
"loss": 0.2402,
|
|
"mean_token_accuracy": 0.9174076914787292,
|
|
"step": 195
|
|
},
|
|
{
|
|
"epoch": 0.9116279069767442,
|
|
"grad_norm": 2.4796955585479736,
|
|
"learning_rate": 1.2363679082908766e-06,
|
|
"loss": 0.2724,
|
|
"mean_token_accuracy": 0.9060444831848145,
|
|
"step": 196
|
|
},
|
|
{
|
|
"epoch": 0.9162790697674419,
|
|
"grad_norm": 3.037348747253418,
|
|
"learning_rate": 1.2135061239363161e-06,
|
|
"loss": 0.2282,
|
|
"mean_token_accuracy": 0.9218890070915222,
|
|
"step": 197
|
|
},
|
|
{
|
|
"epoch": 0.9209302325581395,
|
|
"grad_norm": 2.4801511764526367,
|
|
"learning_rate": 1.1917800765063954e-06,
|
|
"loss": 0.2456,
|
|
"mean_token_accuracy": 0.9177227020263672,
|
|
"step": 198
|
|
},
|
|
{
|
|
"epoch": 0.9255813953488372,
|
|
"grad_norm": 2.8898110389709473,
|
|
"learning_rate": 1.1711955224713209e-06,
|
|
"loss": 0.2776,
|
|
"mean_token_accuracy": 0.9040968418121338,
|
|
"step": 199
|
|
},
|
|
{
|
|
"epoch": 0.9302325581395349,
|
|
"grad_norm": 2.388014078140259,
|
|
"learning_rate": 1.1517579158545386e-06,
|
|
"loss": 0.2513,
|
|
"mean_token_accuracy": 0.911308228969574,
|
|
"step": 200
|
|
},
|
|
{
|
|
"epoch": 0.9348837209302325,
|
|
"grad_norm": 2.6369214057922363,
|
|
"learning_rate": 1.1334724067876463e-06,
|
|
"loss": 0.2835,
|
|
"mean_token_accuracy": 0.9031007885932922,
|
|
"step": 201
|
|
},
|
|
{
|
|
"epoch": 0.9395348837209302,
|
|
"grad_norm": 2.3566486835479736,
|
|
"learning_rate": 1.1163438401458358e-06,
|
|
"loss": 0.2469,
|
|
"mean_token_accuracy": 0.9110381603240967,
|
|
"step": 202
|
|
},
|
|
{
|
|
"epoch": 0.9441860465116279,
|
|
"grad_norm": 2.4852454662323,
|
|
"learning_rate": 1.1003767542642021e-06,
|
|
"loss": 0.2605,
|
|
"mean_token_accuracy": 0.9065656661987305,
|
|
"step": 203
|
|
},
|
|
{
|
|
"epoch": 0.9488372093023256,
|
|
"grad_norm": 2.471855878829956,
|
|
"learning_rate": 1.0855753797352868e-06,
|
|
"loss": 0.2406,
|
|
"mean_token_accuracy": 0.9095502495765686,
|
|
"step": 204
|
|
},
|
|
{
|
|
"epoch": 0.9534883720930233,
|
|
"grad_norm": 2.3165769577026367,
|
|
"learning_rate": 1.0719436382881466e-06,
|
|
"loss": 0.2461,
|
|
"mean_token_accuracy": 0.9113860726356506,
|
|
"step": 205
|
|
},
|
|
{
|
|
"epoch": 0.958139534883721,
|
|
"grad_norm": 2.666945219039917,
|
|
"learning_rate": 1.0594851417492665e-06,
|
|
"loss": 0.2748,
|
|
"mean_token_accuracy": 0.9032517671585083,
|
|
"step": 206
|
|
},
|
|
{
|
|
"epoch": 0.9627906976744186,
|
|
"grad_norm": 2.7581276893615723,
|
|
"learning_rate": 1.0482031910855804e-06,
|
|
"loss": 0.2914,
|
|
"mean_token_accuracy": 0.8979403972625732,
|
|
"step": 207
|
|
},
|
|
{
|
|
"epoch": 0.9674418604651163,
|
|
"grad_norm": 2.7610130310058594,
|
|
"learning_rate": 1.0381007755298547e-06,
|
|
"loss": 0.2683,
|
|
"mean_token_accuracy": 0.9080323576927185,
|
|
"step": 208
|
|
},
|
|
{
|
|
"epoch": 0.9720930232558139,
|
|
"grad_norm": 2.450934648513794,
|
|
"learning_rate": 1.029180571788672e-06,
|
|
"loss": 0.2516,
|
|
"mean_token_accuracy": 0.9152542352676392,
|
|
"step": 209
|
|
},
|
|
{
|
|
"epoch": 0.9767441860465116,
|
|
"grad_norm": 2.3660802841186523,
|
|
"learning_rate": 1.021444943333218e-06,
|
|
"loss": 0.2606,
|
|
"mean_token_accuracy": 0.9119013547897339,
|
|
"step": 210
|
|
},
|
|
{
|
|
"epoch": 0.9813953488372092,
|
|
"grad_norm": 2.59147572517395,
|
|
"learning_rate": 1.0148959397730637e-06,
|
|
"loss": 0.2627,
|
|
"mean_token_accuracy": 0.9101954698562622,
|
|
"step": 211
|
|
},
|
|
{
|
|
"epoch": 0.986046511627907,
|
|
"grad_norm": 2.5691230297088623,
|
|
"learning_rate": 1.0095352963131057e-06,
|
|
"loss": 0.2867,
|
|
"mean_token_accuracy": 0.9024427533149719,
|
|
"step": 212
|
|
},
|
|
{
|
|
"epoch": 0.9906976744186047,
|
|
"grad_norm": 2.443119764328003,
|
|
"learning_rate": 1.0053644332938118e-06,
|
|
"loss": 0.2567,
|
|
"mean_token_accuracy": 0.9110621213912964,
|
|
"step": 213
|
|
},
|
|
{
|
|
"epoch": 0.9953488372093023,
|
|
"grad_norm": 2.7664918899536133,
|
|
"learning_rate": 1.0023844558148912e-06,
|
|
"loss": 0.3081,
|
|
"mean_token_accuracy": 0.896173894405365,
|
|
"step": 214
|
|
},
|
|
{
|
|
"epoch": 1.0,
|
|
"grad_norm": 2.6986541748046875,
|
|
"learning_rate": 1.0005961534424925e-06,
|
|
"loss": 0.1516,
|
|
"mean_token_accuracy": 0.9379363656044006,
|
|
"step": 215
|
|
},
|
|
{
|
|
"epoch": 1.0,
|
|
"step": 215,
|
|
"total_flos": 2.0144468407196058e+17,
|
|
"train_loss": 0.17723823002604552,
|
|
"train_runtime": 1262.477,
|
|
"train_samples_per_second": 5.427,
|
|
"train_steps_per_second": 0.17
|
|
}
|
|
],
|
|
"logging_steps": 1,
|
|
"max_steps": 215,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": true
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 2.0144468407196058e+17,
|
|
"train_batch_size": 4,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|