70e484f736
Model: xd2010/Qwen1.5-MOE-aux-free-sft-math7k-1e-3-gamma-1epoch 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": 19.563091278076172,
|
|
"learning_rate": 0.0,
|
|
"loss": 0.842,
|
|
"mean_token_accuracy": 0.8234314918518066,
|
|
"step": 1
|
|
},
|
|
{
|
|
"epoch": 0.009302325581395349,
|
|
"grad_norm": 21.543142318725586,
|
|
"learning_rate": 4.5454545454545457e-07,
|
|
"loss": 0.8778,
|
|
"mean_token_accuracy": 0.8233309388160706,
|
|
"step": 2
|
|
},
|
|
{
|
|
"epoch": 0.013953488372093023,
|
|
"grad_norm": 19.154624938964844,
|
|
"learning_rate": 9.090909090909091e-07,
|
|
"loss": 0.8504,
|
|
"mean_token_accuracy": 0.8341960310935974,
|
|
"step": 3
|
|
},
|
|
{
|
|
"epoch": 0.018604651162790697,
|
|
"grad_norm": 18.31004524230957,
|
|
"learning_rate": 1.3636363636363636e-06,
|
|
"loss": 0.8171,
|
|
"mean_token_accuracy": 0.8309476971626282,
|
|
"step": 4
|
|
},
|
|
{
|
|
"epoch": 0.023255813953488372,
|
|
"grad_norm": 21.427757263183594,
|
|
"learning_rate": 1.8181818181818183e-06,
|
|
"loss": 0.9323,
|
|
"mean_token_accuracy": 0.810111939907074,
|
|
"step": 5
|
|
},
|
|
{
|
|
"epoch": 0.027906976744186046,
|
|
"grad_norm": 18.453533172607422,
|
|
"learning_rate": 2.2727272727272728e-06,
|
|
"loss": 0.7815,
|
|
"mean_token_accuracy": 0.830949068069458,
|
|
"step": 6
|
|
},
|
|
{
|
|
"epoch": 0.03255813953488372,
|
|
"grad_norm": 17.05266571044922,
|
|
"learning_rate": 2.7272727272727272e-06,
|
|
"loss": 0.8111,
|
|
"mean_token_accuracy": 0.8272174000740051,
|
|
"step": 7
|
|
},
|
|
{
|
|
"epoch": 0.037209302325581395,
|
|
"grad_norm": 10.011034965515137,
|
|
"learning_rate": 3.181818181818182e-06,
|
|
"loss": 0.6374,
|
|
"mean_token_accuracy": 0.8513641953468323,
|
|
"step": 8
|
|
},
|
|
{
|
|
"epoch": 0.04186046511627907,
|
|
"grad_norm": 8.660391807556152,
|
|
"learning_rate": 3.6363636363636366e-06,
|
|
"loss": 0.5639,
|
|
"mean_token_accuracy": 0.8629579544067383,
|
|
"step": 9
|
|
},
|
|
{
|
|
"epoch": 0.046511627906976744,
|
|
"grad_norm": 9.10895824432373,
|
|
"learning_rate": 4.0909090909090915e-06,
|
|
"loss": 0.4929,
|
|
"mean_token_accuracy": 0.8710644841194153,
|
|
"step": 10
|
|
},
|
|
{
|
|
"epoch": 0.05116279069767442,
|
|
"grad_norm": 9.022024154663086,
|
|
"learning_rate": 4.5454545454545455e-06,
|
|
"loss": 0.4222,
|
|
"mean_token_accuracy": 0.8916949033737183,
|
|
"step": 11
|
|
},
|
|
{
|
|
"epoch": 0.05581395348837209,
|
|
"grad_norm": 7.397885322570801,
|
|
"learning_rate": 5e-06,
|
|
"loss": 0.4407,
|
|
"mean_token_accuracy": 0.8812285661697388,
|
|
"step": 12
|
|
},
|
|
{
|
|
"epoch": 0.06046511627906977,
|
|
"grad_norm": 4.204639434814453,
|
|
"learning_rate": 5.4545454545454545e-06,
|
|
"loss": 0.4201,
|
|
"mean_token_accuracy": 0.8759526610374451,
|
|
"step": 13
|
|
},
|
|
{
|
|
"epoch": 0.06511627906976744,
|
|
"grad_norm": 6.326536655426025,
|
|
"learning_rate": 5.90909090909091e-06,
|
|
"loss": 0.366,
|
|
"mean_token_accuracy": 0.8876976370811462,
|
|
"step": 14
|
|
},
|
|
{
|
|
"epoch": 0.06976744186046512,
|
|
"grad_norm": 3.420530080795288,
|
|
"learning_rate": 6.363636363636364e-06,
|
|
"loss": 0.3441,
|
|
"mean_token_accuracy": 0.8966382145881653,
|
|
"step": 15
|
|
},
|
|
{
|
|
"epoch": 0.07441860465116279,
|
|
"grad_norm": 3.6746819019317627,
|
|
"learning_rate": 6.818181818181818e-06,
|
|
"loss": 0.3801,
|
|
"mean_token_accuracy": 0.8873555660247803,
|
|
"step": 16
|
|
},
|
|
{
|
|
"epoch": 0.07906976744186046,
|
|
"grad_norm": 3.0905089378356934,
|
|
"learning_rate": 7.272727272727273e-06,
|
|
"loss": 0.3331,
|
|
"mean_token_accuracy": 0.9028842449188232,
|
|
"step": 17
|
|
},
|
|
{
|
|
"epoch": 0.08372093023255814,
|
|
"grad_norm": 3.0753226280212402,
|
|
"learning_rate": 7.727272727272727e-06,
|
|
"loss": 0.3769,
|
|
"mean_token_accuracy": 0.8833408951759338,
|
|
"step": 18
|
|
},
|
|
{
|
|
"epoch": 0.08837209302325581,
|
|
"grad_norm": 3.367368221282959,
|
|
"learning_rate": 8.181818181818183e-06,
|
|
"loss": 0.3838,
|
|
"mean_token_accuracy": 0.8844019174575806,
|
|
"step": 19
|
|
},
|
|
{
|
|
"epoch": 0.09302325581395349,
|
|
"grad_norm": 2.8647799491882324,
|
|
"learning_rate": 8.636363636363637e-06,
|
|
"loss": 0.3515,
|
|
"mean_token_accuracy": 0.8928460478782654,
|
|
"step": 20
|
|
},
|
|
{
|
|
"epoch": 0.09767441860465116,
|
|
"grad_norm": 3.1543805599212646,
|
|
"learning_rate": 9.090909090909091e-06,
|
|
"loss": 0.4004,
|
|
"mean_token_accuracy": 0.8759110569953918,
|
|
"step": 21
|
|
},
|
|
{
|
|
"epoch": 0.10232558139534884,
|
|
"grad_norm": 3.227837085723877,
|
|
"learning_rate": 9.545454545454547e-06,
|
|
"loss": 0.352,
|
|
"mean_token_accuracy": 0.8826693296432495,
|
|
"step": 22
|
|
},
|
|
{
|
|
"epoch": 0.10697674418604651,
|
|
"grad_norm": 3.01237416267395,
|
|
"learning_rate": 1e-05,
|
|
"loss": 0.3455,
|
|
"mean_token_accuracy": 0.8948639631271362,
|
|
"step": 23
|
|
},
|
|
{
|
|
"epoch": 0.11162790697674418,
|
|
"grad_norm": 3.248323440551758,
|
|
"learning_rate": 9.999403846557509e-06,
|
|
"loss": 0.3715,
|
|
"mean_token_accuracy": 0.8808307647705078,
|
|
"step": 24
|
|
},
|
|
{
|
|
"epoch": 0.11627906976744186,
|
|
"grad_norm": 3.0108933448791504,
|
|
"learning_rate": 9.99761554418511e-06,
|
|
"loss": 0.3335,
|
|
"mean_token_accuracy": 0.8970856070518494,
|
|
"step": 25
|
|
},
|
|
{
|
|
"epoch": 0.12093023255813953,
|
|
"grad_norm": 3.032703399658203,
|
|
"learning_rate": 9.99463556670619e-06,
|
|
"loss": 0.3349,
|
|
"mean_token_accuracy": 0.8989508748054504,
|
|
"step": 26
|
|
},
|
|
{
|
|
"epoch": 0.12558139534883722,
|
|
"grad_norm": 3.1234946250915527,
|
|
"learning_rate": 9.990464703686895e-06,
|
|
"loss": 0.3647,
|
|
"mean_token_accuracy": 0.8929548859596252,
|
|
"step": 27
|
|
},
|
|
{
|
|
"epoch": 0.13023255813953488,
|
|
"grad_norm": 3.0966460704803467,
|
|
"learning_rate": 9.985104060226937e-06,
|
|
"loss": 0.3527,
|
|
"mean_token_accuracy": 0.8932457566261292,
|
|
"step": 28
|
|
},
|
|
{
|
|
"epoch": 0.13488372093023257,
|
|
"grad_norm": 2.9237380027770996,
|
|
"learning_rate": 9.978555056666784e-06,
|
|
"loss": 0.3755,
|
|
"mean_token_accuracy": 0.878600537776947,
|
|
"step": 29
|
|
},
|
|
{
|
|
"epoch": 0.13953488372093023,
|
|
"grad_norm": 3.067922830581665,
|
|
"learning_rate": 9.97081942821133e-06,
|
|
"loss": 0.3706,
|
|
"mean_token_accuracy": 0.8867707252502441,
|
|
"step": 30
|
|
},
|
|
{
|
|
"epoch": 0.14418604651162792,
|
|
"grad_norm": 3.369419574737549,
|
|
"learning_rate": 9.961899224470146e-06,
|
|
"loss": 0.3949,
|
|
"mean_token_accuracy": 0.8826349377632141,
|
|
"step": 31
|
|
},
|
|
{
|
|
"epoch": 0.14883720930232558,
|
|
"grad_norm": 3.1018178462982178,
|
|
"learning_rate": 9.95179680891442e-06,
|
|
"loss": 0.3196,
|
|
"mean_token_accuracy": 0.90621018409729,
|
|
"step": 32
|
|
},
|
|
{
|
|
"epoch": 0.15348837209302327,
|
|
"grad_norm": 3.642944812774658,
|
|
"learning_rate": 9.940514858250736e-06,
|
|
"loss": 0.3547,
|
|
"mean_token_accuracy": 0.8922168612480164,
|
|
"step": 33
|
|
},
|
|
{
|
|
"epoch": 0.15813953488372093,
|
|
"grad_norm": 2.888031244277954,
|
|
"learning_rate": 9.928056361711854e-06,
|
|
"loss": 0.3032,
|
|
"mean_token_accuracy": 0.903816819190979,
|
|
"step": 34
|
|
},
|
|
{
|
|
"epoch": 0.16279069767441862,
|
|
"grad_norm": 2.99672794342041,
|
|
"learning_rate": 9.914424620264714e-06,
|
|
"loss": 0.3585,
|
|
"mean_token_accuracy": 0.8923381567001343,
|
|
"step": 35
|
|
},
|
|
{
|
|
"epoch": 0.16744186046511628,
|
|
"grad_norm": 3.023254632949829,
|
|
"learning_rate": 9.899623245735798e-06,
|
|
"loss": 0.3284,
|
|
"mean_token_accuracy": 0.8969854116439819,
|
|
"step": 36
|
|
},
|
|
{
|
|
"epoch": 0.17209302325581396,
|
|
"grad_norm": 2.9886157512664795,
|
|
"learning_rate": 9.883656159854166e-06,
|
|
"loss": 0.3175,
|
|
"mean_token_accuracy": 0.9026827216148376,
|
|
"step": 37
|
|
},
|
|
{
|
|
"epoch": 0.17674418604651163,
|
|
"grad_norm": 3.0132319927215576,
|
|
"learning_rate": 9.866527593212355e-06,
|
|
"loss": 0.3321,
|
|
"mean_token_accuracy": 0.8988155126571655,
|
|
"step": 38
|
|
},
|
|
{
|
|
"epoch": 0.1813953488372093,
|
|
"grad_norm": 2.979572296142578,
|
|
"learning_rate": 9.848242084145462e-06,
|
|
"loss": 0.3195,
|
|
"mean_token_accuracy": 0.9020941257476807,
|
|
"step": 39
|
|
},
|
|
{
|
|
"epoch": 0.18604651162790697,
|
|
"grad_norm": 2.8878800868988037,
|
|
"learning_rate": 9.82880447752868e-06,
|
|
"loss": 0.3379,
|
|
"mean_token_accuracy": 0.892432451248169,
|
|
"step": 40
|
|
},
|
|
{
|
|
"epoch": 0.19069767441860466,
|
|
"grad_norm": 2.995546817779541,
|
|
"learning_rate": 9.808219923493606e-06,
|
|
"loss": 0.3316,
|
|
"mean_token_accuracy": 0.8975328803062439,
|
|
"step": 41
|
|
},
|
|
{
|
|
"epoch": 0.19534883720930232,
|
|
"grad_norm": 3.1957032680511475,
|
|
"learning_rate": 9.786493876063685e-06,
|
|
"loss": 0.3289,
|
|
"mean_token_accuracy": 0.8968408703804016,
|
|
"step": 42
|
|
},
|
|
{
|
|
"epoch": 0.2,
|
|
"grad_norm": 3.114238977432251,
|
|
"learning_rate": 9.763632091709125e-06,
|
|
"loss": 0.3094,
|
|
"mean_token_accuracy": 0.8965780735015869,
|
|
"step": 43
|
|
},
|
|
{
|
|
"epoch": 0.20465116279069767,
|
|
"grad_norm": 2.9550487995147705,
|
|
"learning_rate": 9.739640627821678e-06,
|
|
"loss": 0.3372,
|
|
"mean_token_accuracy": 0.898785412311554,
|
|
"step": 44
|
|
},
|
|
{
|
|
"epoch": 0.20930232558139536,
|
|
"grad_norm": 3.2309701442718506,
|
|
"learning_rate": 9.714525841109697e-06,
|
|
"loss": 0.2948,
|
|
"mean_token_accuracy": 0.9049268960952759,
|
|
"step": 45
|
|
},
|
|
{
|
|
"epoch": 0.21395348837209302,
|
|
"grad_norm": 3.1183102130889893,
|
|
"learning_rate": 9.68829438591387e-06,
|
|
"loss": 0.3041,
|
|
"mean_token_accuracy": 0.9000358581542969,
|
|
"step": 46
|
|
},
|
|
{
|
|
"epoch": 0.2186046511627907,
|
|
"grad_norm": 3.0251879692077637,
|
|
"learning_rate": 9.660953212444116e-06,
|
|
"loss": 0.3135,
|
|
"mean_token_accuracy": 0.8976897597312927,
|
|
"step": 47
|
|
},
|
|
{
|
|
"epoch": 0.22325581395348837,
|
|
"grad_norm": 2.8811988830566406,
|
|
"learning_rate": 9.632509564938073e-06,
|
|
"loss": 0.3213,
|
|
"mean_token_accuracy": 0.8950562477111816,
|
|
"step": 48
|
|
},
|
|
{
|
|
"epoch": 0.22790697674418606,
|
|
"grad_norm": 2.8973190784454346,
|
|
"learning_rate": 9.60297097974169e-06,
|
|
"loss": 0.3134,
|
|
"mean_token_accuracy": 0.8969748020172119,
|
|
"step": 49
|
|
},
|
|
{
|
|
"epoch": 0.23255813953488372,
|
|
"grad_norm": 2.8434340953826904,
|
|
"learning_rate": 9.572345283312407e-06,
|
|
"loss": 0.3161,
|
|
"mean_token_accuracy": 0.8976973295211792,
|
|
"step": 50
|
|
},
|
|
{
|
|
"epoch": 0.2372093023255814,
|
|
"grad_norm": 3.1915981769561768,
|
|
"learning_rate": 9.540640590145496e-06,
|
|
"loss": 0.3334,
|
|
"mean_token_accuracy": 0.88165682554245,
|
|
"step": 51
|
|
},
|
|
{
|
|
"epoch": 0.24186046511627907,
|
|
"grad_norm": 2.863004684448242,
|
|
"learning_rate": 9.507865300624057e-06,
|
|
"loss": 0.3354,
|
|
"mean_token_accuracy": 0.8871039748191833,
|
|
"step": 52
|
|
},
|
|
{
|
|
"epoch": 0.24651162790697675,
|
|
"grad_norm": 2.9029958248138428,
|
|
"learning_rate": 9.474028098793277e-06,
|
|
"loss": 0.3121,
|
|
"mean_token_accuracy": 0.8972073793411255,
|
|
"step": 53
|
|
},
|
|
{
|
|
"epoch": 0.25116279069767444,
|
|
"grad_norm": 2.8073384761810303,
|
|
"learning_rate": 9.439137950059539e-06,
|
|
"loss": 0.3275,
|
|
"mean_token_accuracy": 0.8906416296958923,
|
|
"step": 54
|
|
},
|
|
{
|
|
"epoch": 0.2558139534883721,
|
|
"grad_norm": 2.7467398643493652,
|
|
"learning_rate": 9.403204098814965e-06,
|
|
"loss": 0.3281,
|
|
"mean_token_accuracy": 0.8906298279762268,
|
|
"step": 55
|
|
},
|
|
{
|
|
"epoch": 0.26046511627906976,
|
|
"grad_norm": 2.7916853427886963,
|
|
"learning_rate": 9.366236065988053e-06,
|
|
"loss": 0.3221,
|
|
"mean_token_accuracy": 0.8957247138023376,
|
|
"step": 56
|
|
},
|
|
{
|
|
"epoch": 0.2651162790697674,
|
|
"grad_norm": 2.3894519805908203,
|
|
"learning_rate": 9.32824364652104e-06,
|
|
"loss": 0.2809,
|
|
"mean_token_accuracy": 0.9081918597221375,
|
|
"step": 57
|
|
},
|
|
{
|
|
"epoch": 0.26976744186046514,
|
|
"grad_norm": 2.4881339073181152,
|
|
"learning_rate": 9.289236906774663e-06,
|
|
"loss": 0.3101,
|
|
"mean_token_accuracy": 0.8962963223457336,
|
|
"step": 58
|
|
},
|
|
{
|
|
"epoch": 0.2744186046511628,
|
|
"grad_norm": 2.7378764152526855,
|
|
"learning_rate": 9.249226181861e-06,
|
|
"loss": 0.2909,
|
|
"mean_token_accuracy": 0.9027132987976074,
|
|
"step": 59
|
|
},
|
|
{
|
|
"epoch": 0.27906976744186046,
|
|
"grad_norm": 2.5940091609954834,
|
|
"learning_rate": 9.208222072905113e-06,
|
|
"loss": 0.3071,
|
|
"mean_token_accuracy": 0.8917708396911621,
|
|
"step": 60
|
|
},
|
|
{
|
|
"epoch": 0.2837209302325581,
|
|
"grad_norm": 2.662109613418579,
|
|
"learning_rate": 9.166235444236209e-06,
|
|
"loss": 0.3409,
|
|
"mean_token_accuracy": 0.8838131427764893,
|
|
"step": 61
|
|
},
|
|
{
|
|
"epoch": 0.28837209302325584,
|
|
"grad_norm": 2.6637372970581055,
|
|
"learning_rate": 9.123277420509053e-06,
|
|
"loss": 0.3341,
|
|
"mean_token_accuracy": 0.8877442479133606,
|
|
"step": 62
|
|
},
|
|
{
|
|
"epoch": 0.2930232558139535,
|
|
"grad_norm": 2.4996681213378906,
|
|
"learning_rate": 9.079359383756411e-06,
|
|
"loss": 0.312,
|
|
"mean_token_accuracy": 0.8987603187561035,
|
|
"step": 63
|
|
},
|
|
{
|
|
"epoch": 0.29767441860465116,
|
|
"grad_norm": 2.4879515171051025,
|
|
"learning_rate": 9.034492970373305e-06,
|
|
"loss": 0.3065,
|
|
"mean_token_accuracy": 0.8970829248428345,
|
|
"step": 64
|
|
},
|
|
{
|
|
"epoch": 0.3023255813953488,
|
|
"grad_norm": 2.618191719055176,
|
|
"learning_rate": 8.988690068033864e-06,
|
|
"loss": 0.3333,
|
|
"mean_token_accuracy": 0.8905907869338989,
|
|
"step": 65
|
|
},
|
|
{
|
|
"epoch": 0.30697674418604654,
|
|
"grad_norm": 2.3061347007751465,
|
|
"learning_rate": 8.941962812541604e-06,
|
|
"loss": 0.2995,
|
|
"mean_token_accuracy": 0.8940735459327698,
|
|
"step": 66
|
|
},
|
|
{
|
|
"epoch": 0.3116279069767442,
|
|
"grad_norm": 2.4990477561950684,
|
|
"learning_rate": 8.894323584613951e-06,
|
|
"loss": 0.319,
|
|
"mean_token_accuracy": 0.888198733329773,
|
|
"step": 67
|
|
},
|
|
{
|
|
"epoch": 0.31627906976744186,
|
|
"grad_norm": 2.426720142364502,
|
|
"learning_rate": 8.845785006601898e-06,
|
|
"loss": 0.2816,
|
|
"mean_token_accuracy": 0.9047619104385376,
|
|
"step": 68
|
|
},
|
|
{
|
|
"epoch": 0.3209302325581395,
|
|
"grad_norm": 2.5082154273986816,
|
|
"learning_rate": 8.796359939145614e-06,
|
|
"loss": 0.3077,
|
|
"mean_token_accuracy": 0.8942273855209351,
|
|
"step": 69
|
|
},
|
|
{
|
|
"epoch": 0.32558139534883723,
|
|
"grad_norm": 2.6549720764160156,
|
|
"learning_rate": 8.74606147776692e-06,
|
|
"loss": 0.3093,
|
|
"mean_token_accuracy": 0.8931810259819031,
|
|
"step": 70
|
|
},
|
|
{
|
|
"epoch": 0.3302325581395349,
|
|
"grad_norm": 2.528726577758789,
|
|
"learning_rate": 8.694902949399555e-06,
|
|
"loss": 0.3084,
|
|
"mean_token_accuracy": 0.8978523015975952,
|
|
"step": 71
|
|
},
|
|
{
|
|
"epoch": 0.33488372093023255,
|
|
"grad_norm": 2.5078654289245605,
|
|
"learning_rate": 8.642897908858096e-06,
|
|
"loss": 0.3105,
|
|
"mean_token_accuracy": 0.8962626457214355,
|
|
"step": 72
|
|
},
|
|
{
|
|
"epoch": 0.3395348837209302,
|
|
"grad_norm": 2.638470411300659,
|
|
"learning_rate": 8.590060135246516e-06,
|
|
"loss": 0.3341,
|
|
"mean_token_accuracy": 0.8825904130935669,
|
|
"step": 73
|
|
},
|
|
{
|
|
"epoch": 0.34418604651162793,
|
|
"grad_norm": 2.316749334335327,
|
|
"learning_rate": 8.53640362830732e-06,
|
|
"loss": 0.2888,
|
|
"mean_token_accuracy": 0.9002068042755127,
|
|
"step": 74
|
|
},
|
|
{
|
|
"epoch": 0.3488372093023256,
|
|
"grad_norm": 2.6540651321411133,
|
|
"learning_rate": 8.481942604712209e-06,
|
|
"loss": 0.3172,
|
|
"mean_token_accuracy": 0.8934825658798218,
|
|
"step": 75
|
|
},
|
|
{
|
|
"epoch": 0.35348837209302325,
|
|
"grad_norm": 2.3482789993286133,
|
|
"learning_rate": 8.426691494295269e-06,
|
|
"loss": 0.2869,
|
|
"mean_token_accuracy": 0.9042777419090271,
|
|
"step": 76
|
|
},
|
|
{
|
|
"epoch": 0.3581395348837209,
|
|
"grad_norm": 2.464599609375,
|
|
"learning_rate": 8.370664936229688e-06,
|
|
"loss": 0.3122,
|
|
"mean_token_accuracy": 0.8950356841087341,
|
|
"step": 77
|
|
},
|
|
{
|
|
"epoch": 0.3627906976744186,
|
|
"grad_norm": 2.4648303985595703,
|
|
"learning_rate": 8.313877775149009e-06,
|
|
"loss": 0.2732,
|
|
"mean_token_accuracy": 0.9056981205940247,
|
|
"step": 78
|
|
},
|
|
{
|
|
"epoch": 0.3674418604651163,
|
|
"grad_norm": 2.753995418548584,
|
|
"learning_rate": 8.256345057213925e-06,
|
|
"loss": 0.3128,
|
|
"mean_token_accuracy": 0.8913813233375549,
|
|
"step": 79
|
|
},
|
|
{
|
|
"epoch": 0.37209302325581395,
|
|
"grad_norm": 2.361708879470825,
|
|
"learning_rate": 8.198082026125707e-06,
|
|
"loss": 0.3283,
|
|
"mean_token_accuracy": 0.8879844546318054,
|
|
"step": 80
|
|
},
|
|
{
|
|
"epoch": 0.3767441860465116,
|
|
"grad_norm": 2.556525707244873,
|
|
"learning_rate": 8.139104119087265e-06,
|
|
"loss": 0.2841,
|
|
"mean_token_accuracy": 0.9036663174629211,
|
|
"step": 81
|
|
},
|
|
{
|
|
"epoch": 0.3813953488372093,
|
|
"grad_norm": 2.482909917831421,
|
|
"learning_rate": 8.07942696271296e-06,
|
|
"loss": 0.2808,
|
|
"mean_token_accuracy": 0.9009503126144409,
|
|
"step": 82
|
|
},
|
|
{
|
|
"epoch": 0.386046511627907,
|
|
"grad_norm": 2.3375966548919678,
|
|
"learning_rate": 8.019066368888222e-06,
|
|
"loss": 0.2867,
|
|
"mean_token_accuracy": 0.9043236970901489,
|
|
"step": 83
|
|
},
|
|
{
|
|
"epoch": 0.39069767441860465,
|
|
"grad_norm": 2.4593124389648438,
|
|
"learning_rate": 7.958038330580067e-06,
|
|
"loss": 0.3289,
|
|
"mean_token_accuracy": 0.8837209343910217,
|
|
"step": 84
|
|
},
|
|
{
|
|
"epoch": 0.3953488372093023,
|
|
"grad_norm": 2.3785974979400635,
|
|
"learning_rate": 7.89635901759967e-06,
|
|
"loss": 0.2974,
|
|
"mean_token_accuracy": 0.8938296437263489,
|
|
"step": 85
|
|
},
|
|
{
|
|
"epoch": 0.4,
|
|
"grad_norm": 2.7969272136688232,
|
|
"learning_rate": 7.834044772318033e-06,
|
|
"loss": 0.3279,
|
|
"mean_token_accuracy": 0.8935073018074036,
|
|
"step": 86
|
|
},
|
|
{
|
|
"epoch": 0.4046511627906977,
|
|
"grad_norm": 2.5054378509521484,
|
|
"learning_rate": 7.77111210533597e-06,
|
|
"loss": 0.3006,
|
|
"mean_token_accuracy": 0.8904744982719421,
|
|
"step": 87
|
|
},
|
|
{
|
|
"epoch": 0.40930232558139534,
|
|
"grad_norm": 2.8241288661956787,
|
|
"learning_rate": 7.707577691109519e-06,
|
|
"loss": 0.3572,
|
|
"mean_token_accuracy": 0.8787984251976013,
|
|
"step": 88
|
|
},
|
|
{
|
|
"epoch": 0.413953488372093,
|
|
"grad_norm": 3.031515121459961,
|
|
"learning_rate": 7.6434583635319e-06,
|
|
"loss": 0.2999,
|
|
"mean_token_accuracy": 0.904349148273468,
|
|
"step": 89
|
|
},
|
|
{
|
|
"epoch": 0.4186046511627907,
|
|
"grad_norm": 2.243628978729248,
|
|
"learning_rate": 7.578771111473276e-06,
|
|
"loss": 0.2569,
|
|
"mean_token_accuracy": 0.9113546013832092,
|
|
"step": 90
|
|
},
|
|
{
|
|
"epoch": 0.4232558139534884,
|
|
"grad_norm": 2.393187999725342,
|
|
"learning_rate": 7.513533074279427e-06,
|
|
"loss": 0.2673,
|
|
"mean_token_accuracy": 0.9085960984230042,
|
|
"step": 91
|
|
},
|
|
{
|
|
"epoch": 0.42790697674418604,
|
|
"grad_norm": 2.4981424808502197,
|
|
"learning_rate": 7.4477615372305545e-06,
|
|
"loss": 0.3206,
|
|
"mean_token_accuracy": 0.8938068151473999,
|
|
"step": 92
|
|
},
|
|
{
|
|
"epoch": 0.4325581395348837,
|
|
"grad_norm": 2.3902385234832764,
|
|
"learning_rate": 7.3814739269614265e-06,
|
|
"loss": 0.2927,
|
|
"mean_token_accuracy": 0.8991581201553345,
|
|
"step": 93
|
|
},
|
|
{
|
|
"epoch": 0.4372093023255814,
|
|
"grad_norm": 2.3885436058044434,
|
|
"learning_rate": 7.314687806844067e-06,
|
|
"loss": 0.3055,
|
|
"mean_token_accuracy": 0.8974854946136475,
|
|
"step": 94
|
|
},
|
|
{
|
|
"epoch": 0.4418604651162791,
|
|
"grad_norm": 2.4455854892730713,
|
|
"learning_rate": 7.247420872334221e-06,
|
|
"loss": 0.3022,
|
|
"mean_token_accuracy": 0.8960738778114319,
|
|
"step": 95
|
|
},
|
|
{
|
|
"epoch": 0.44651162790697674,
|
|
"grad_norm": 2.212984323501587,
|
|
"learning_rate": 7.179690946282808e-06,
|
|
"loss": 0.2798,
|
|
"mean_token_accuracy": 0.9022396206855774,
|
|
"step": 96
|
|
},
|
|
{
|
|
"epoch": 0.4511627906976744,
|
|
"grad_norm": 2.587747573852539,
|
|
"learning_rate": 7.111515974213639e-06,
|
|
"loss": 0.2468,
|
|
"mean_token_accuracy": 0.9119572639465332,
|
|
"step": 97
|
|
},
|
|
{
|
|
"epoch": 0.4558139534883721,
|
|
"grad_norm": 2.3034160137176514,
|
|
"learning_rate": 7.042914019568621e-06,
|
|
"loss": 0.3013,
|
|
"mean_token_accuracy": 0.8919404149055481,
|
|
"step": 98
|
|
},
|
|
{
|
|
"epoch": 0.4604651162790698,
|
|
"grad_norm": 2.1824116706848145,
|
|
"learning_rate": 6.973903258921719e-06,
|
|
"loss": 0.2548,
|
|
"mean_token_accuracy": 0.9117500185966492,
|
|
"step": 99
|
|
},
|
|
{
|
|
"epoch": 0.46511627906976744,
|
|
"grad_norm": 2.3703949451446533,
|
|
"learning_rate": 6.904501977162949e-06,
|
|
"loss": 0.3254,
|
|
"mean_token_accuracy": 0.8863078951835632,
|
|
"step": 100
|
|
},
|
|
{
|
|
"epoch": 0.4697674418604651,
|
|
"grad_norm": 2.3646938800811768,
|
|
"learning_rate": 6.834728562653659e-06,
|
|
"loss": 0.3082,
|
|
"mean_token_accuracy": 0.8913007974624634,
|
|
"step": 101
|
|
},
|
|
{
|
|
"epoch": 0.4744186046511628,
|
|
"grad_norm": 2.4611222743988037,
|
|
"learning_rate": 6.764601502354403e-06,
|
|
"loss": 0.3038,
|
|
"mean_token_accuracy": 0.8966363072395325,
|
|
"step": 102
|
|
},
|
|
{
|
|
"epoch": 0.4790697674418605,
|
|
"grad_norm": 2.223982334136963,
|
|
"learning_rate": 6.6941393769266995e-06,
|
|
"loss": 0.2949,
|
|
"mean_token_accuracy": 0.9001782536506653,
|
|
"step": 103
|
|
},
|
|
{
|
|
"epoch": 0.48372093023255813,
|
|
"grad_norm": 2.121941089630127,
|
|
"learning_rate": 6.6233608558099405e-06,
|
|
"loss": 0.275,
|
|
"mean_token_accuracy": 0.906988263130188,
|
|
"step": 104
|
|
},
|
|
{
|
|
"epoch": 0.4883720930232558,
|
|
"grad_norm": 2.263637065887451,
|
|
"learning_rate": 6.552284692274803e-06,
|
|
"loss": 0.3072,
|
|
"mean_token_accuracy": 0.891097903251648,
|
|
"step": 105
|
|
},
|
|
{
|
|
"epoch": 0.4930232558139535,
|
|
"grad_norm": 2.3840157985687256,
|
|
"learning_rate": 6.48092971845443e-06,
|
|
"loss": 0.2792,
|
|
"mean_token_accuracy": 0.9033563733100891,
|
|
"step": 106
|
|
},
|
|
{
|
|
"epoch": 0.49767441860465117,
|
|
"grad_norm": 2.4557018280029297,
|
|
"learning_rate": 6.409314840354724e-06,
|
|
"loss": 0.3368,
|
|
"mean_token_accuracy": 0.8887357711791992,
|
|
"step": 107
|
|
},
|
|
{
|
|
"epoch": 0.5023255813953489,
|
|
"grad_norm": 2.482020378112793,
|
|
"learning_rate": 6.337459032845068e-06,
|
|
"loss": 0.3139,
|
|
"mean_token_accuracy": 0.888927161693573,
|
|
"step": 108
|
|
},
|
|
{
|
|
"epoch": 0.5069767441860465,
|
|
"grad_norm": 3.620486259460449,
|
|
"learning_rate": 6.2653813346308e-06,
|
|
"loss": 0.2704,
|
|
"mean_token_accuracy": 0.9052498936653137,
|
|
"step": 109
|
|
},
|
|
{
|
|
"epoch": 0.5116279069767442,
|
|
"grad_norm": 2.3240528106689453,
|
|
"learning_rate": 6.193100843208772e-06,
|
|
"loss": 0.2815,
|
|
"mean_token_accuracy": 0.8988596200942993,
|
|
"step": 110
|
|
},
|
|
{
|
|
"epoch": 0.5162790697674419,
|
|
"grad_norm": 2.380955219268799,
|
|
"learning_rate": 6.120636709807334e-06,
|
|
"loss": 0.3092,
|
|
"mean_token_accuracy": 0.8890586495399475,
|
|
"step": 111
|
|
},
|
|
{
|
|
"epoch": 0.5209302325581395,
|
|
"grad_norm": 2.4002912044525146,
|
|
"learning_rate": 6.048008134312078e-06,
|
|
"loss": 0.2879,
|
|
"mean_token_accuracy": 0.9019830226898193,
|
|
"step": 112
|
|
},
|
|
{
|
|
"epoch": 0.5255813953488372,
|
|
"grad_norm": 2.3931264877319336,
|
|
"learning_rate": 5.975234360178698e-06,
|
|
"loss": 0.3253,
|
|
"mean_token_accuracy": 0.8930376172065735,
|
|
"step": 113
|
|
},
|
|
{
|
|
"epoch": 0.5302325581395348,
|
|
"grad_norm": 2.1816189289093018,
|
|
"learning_rate": 5.902334669334287e-06,
|
|
"loss": 0.2981,
|
|
"mean_token_accuracy": 0.9001047611236572,
|
|
"step": 114
|
|
},
|
|
{
|
|
"epoch": 0.5348837209302325,
|
|
"grad_norm": 2.4347808361053467,
|
|
"learning_rate": 5.829328377068476e-06,
|
|
"loss": 0.295,
|
|
"mean_token_accuracy": 0.8933088779449463,
|
|
"step": 115
|
|
},
|
|
{
|
|
"epoch": 0.5395348837209303,
|
|
"grad_norm": 2.4828367233276367,
|
|
"learning_rate": 5.756234826915686e-06,
|
|
"loss": 0.2991,
|
|
"mean_token_accuracy": 0.8999999761581421,
|
|
"step": 116
|
|
},
|
|
{
|
|
"epoch": 0.5441860465116279,
|
|
"grad_norm": 2.5322439670562744,
|
|
"learning_rate": 5.683073385529938e-06,
|
|
"loss": 0.3114,
|
|
"mean_token_accuracy": 0.8893616795539856,
|
|
"step": 117
|
|
},
|
|
{
|
|
"epoch": 0.5488372093023256,
|
|
"grad_norm": 2.2047011852264404,
|
|
"learning_rate": 5.60986343755352e-06,
|
|
"loss": 0.2613,
|
|
"mean_token_accuracy": 0.9094377160072327,
|
|
"step": 118
|
|
},
|
|
{
|
|
"epoch": 0.5534883720930233,
|
|
"grad_norm": 2.483036994934082,
|
|
"learning_rate": 5.536624380480878e-06,
|
|
"loss": 0.2511,
|
|
"mean_token_accuracy": 0.9123600125312805,
|
|
"step": 119
|
|
},
|
|
{
|
|
"epoch": 0.5581395348837209,
|
|
"grad_norm": 2.2880711555480957,
|
|
"learning_rate": 5.4633756195191235e-06,
|
|
"loss": 0.2587,
|
|
"mean_token_accuracy": 0.9095374941825867,
|
|
"step": 120
|
|
},
|
|
{
|
|
"epoch": 0.5627906976744186,
|
|
"grad_norm": 2.0043041706085205,
|
|
"learning_rate": 5.390136562446482e-06,
|
|
"loss": 0.2257,
|
|
"mean_token_accuracy": 0.9178202152252197,
|
|
"step": 121
|
|
},
|
|
{
|
|
"epoch": 0.5674418604651162,
|
|
"grad_norm": 2.4865000247955322,
|
|
"learning_rate": 5.316926614470063e-06,
|
|
"loss": 0.3032,
|
|
"mean_token_accuracy": 0.8945955634117126,
|
|
"step": 122
|
|
},
|
|
{
|
|
"epoch": 0.5720930232558139,
|
|
"grad_norm": 2.6121153831481934,
|
|
"learning_rate": 5.2437651730843165e-06,
|
|
"loss": 0.3357,
|
|
"mean_token_accuracy": 0.8777977228164673,
|
|
"step": 123
|
|
},
|
|
{
|
|
"epoch": 0.5767441860465117,
|
|
"grad_norm": 2.5301575660705566,
|
|
"learning_rate": 5.170671622931527e-06,
|
|
"loss": 0.309,
|
|
"mean_token_accuracy": 0.8922097086906433,
|
|
"step": 124
|
|
},
|
|
{
|
|
"epoch": 0.5813953488372093,
|
|
"grad_norm": 2.0600168704986572,
|
|
"learning_rate": 5.097665330665714e-06,
|
|
"loss": 0.2362,
|
|
"mean_token_accuracy": 0.916695237159729,
|
|
"step": 125
|
|
},
|
|
{
|
|
"epoch": 0.586046511627907,
|
|
"grad_norm": 2.6038460731506348,
|
|
"learning_rate": 5.024765639821305e-06,
|
|
"loss": 0.3186,
|
|
"mean_token_accuracy": 0.8942528963088989,
|
|
"step": 126
|
|
},
|
|
{
|
|
"epoch": 0.5906976744186047,
|
|
"grad_norm": 2.052591323852539,
|
|
"learning_rate": 4.951991865687923e-06,
|
|
"loss": 0.2498,
|
|
"mean_token_accuracy": 0.9167211055755615,
|
|
"step": 127
|
|
},
|
|
{
|
|
"epoch": 0.5953488372093023,
|
|
"grad_norm": 2.366655111312866,
|
|
"learning_rate": 4.879363290192667e-06,
|
|
"loss": 0.294,
|
|
"mean_token_accuracy": 0.9024970531463623,
|
|
"step": 128
|
|
},
|
|
{
|
|
"epoch": 0.6,
|
|
"grad_norm": 2.7266809940338135,
|
|
"learning_rate": 4.806899156791231e-06,
|
|
"loss": 0.3559,
|
|
"mean_token_accuracy": 0.8837209343910217,
|
|
"step": 129
|
|
},
|
|
{
|
|
"epoch": 0.6046511627906976,
|
|
"grad_norm": 2.3466830253601074,
|
|
"learning_rate": 4.734618665369202e-06,
|
|
"loss": 0.2796,
|
|
"mean_token_accuracy": 0.8968087434768677,
|
|
"step": 130
|
|
},
|
|
{
|
|
"epoch": 0.6093023255813953,
|
|
"grad_norm": 2.336235523223877,
|
|
"learning_rate": 4.662540967154934e-06,
|
|
"loss": 0.2652,
|
|
"mean_token_accuracy": 0.906707227230072,
|
|
"step": 131
|
|
},
|
|
{
|
|
"epoch": 0.6139534883720931,
|
|
"grad_norm": 2.1116087436676025,
|
|
"learning_rate": 4.5906851596452765e-06,
|
|
"loss": 0.2681,
|
|
"mean_token_accuracy": 0.9023987054824829,
|
|
"step": 132
|
|
},
|
|
{
|
|
"epoch": 0.6186046511627907,
|
|
"grad_norm": 2.292062520980835,
|
|
"learning_rate": 4.519070281545571e-06,
|
|
"loss": 0.3037,
|
|
"mean_token_accuracy": 0.8935209512710571,
|
|
"step": 133
|
|
},
|
|
{
|
|
"epoch": 0.6232558139534884,
|
|
"grad_norm": 2.258714199066162,
|
|
"learning_rate": 4.447715307725197e-06,
|
|
"loss": 0.3077,
|
|
"mean_token_accuracy": 0.8988802433013916,
|
|
"step": 134
|
|
},
|
|
{
|
|
"epoch": 0.627906976744186,
|
|
"grad_norm": 2.3175172805786133,
|
|
"learning_rate": 4.376639144190061e-06,
|
|
"loss": 0.2984,
|
|
"mean_token_accuracy": 0.8972256183624268,
|
|
"step": 135
|
|
},
|
|
{
|
|
"epoch": 0.6325581395348837,
|
|
"grad_norm": 2.546627998352051,
|
|
"learning_rate": 4.305860623073304e-06,
|
|
"loss": 0.2951,
|
|
"mean_token_accuracy": 0.8994247913360596,
|
|
"step": 136
|
|
},
|
|
{
|
|
"epoch": 0.6372093023255814,
|
|
"grad_norm": 2.150290012359619,
|
|
"learning_rate": 4.2353984976456e-06,
|
|
"loss": 0.2789,
|
|
"mean_token_accuracy": 0.9035775065422058,
|
|
"step": 137
|
|
},
|
|
{
|
|
"epoch": 0.641860465116279,
|
|
"grad_norm": 2.1709845066070557,
|
|
"learning_rate": 4.1652714373463435e-06,
|
|
"loss": 0.2861,
|
|
"mean_token_accuracy": 0.8999999761581421,
|
|
"step": 138
|
|
},
|
|
{
|
|
"epoch": 0.6465116279069767,
|
|
"grad_norm": 2.230078935623169,
|
|
"learning_rate": 4.095498022837051e-06,
|
|
"loss": 0.2718,
|
|
"mean_token_accuracy": 0.9082063436508179,
|
|
"step": 139
|
|
},
|
|
{
|
|
"epoch": 0.6511627906976745,
|
|
"grad_norm": 2.339871406555176,
|
|
"learning_rate": 4.026096741078281e-06,
|
|
"loss": 0.3107,
|
|
"mean_token_accuracy": 0.8929083943367004,
|
|
"step": 140
|
|
},
|
|
{
|
|
"epoch": 0.6558139534883721,
|
|
"grad_norm": 2.299539566040039,
|
|
"learning_rate": 3.957085980431382e-06,
|
|
"loss": 0.3036,
|
|
"mean_token_accuracy": 0.8962635397911072,
|
|
"step": 141
|
|
},
|
|
{
|
|
"epoch": 0.6604651162790698,
|
|
"grad_norm": 2.1768903732299805,
|
|
"learning_rate": 3.888484025786364e-06,
|
|
"loss": 0.2767,
|
|
"mean_token_accuracy": 0.9040858745574951,
|
|
"step": 142
|
|
},
|
|
{
|
|
"epoch": 0.6651162790697674,
|
|
"grad_norm": 2.09299373626709,
|
|
"learning_rate": 3.820309053717195e-06,
|
|
"loss": 0.2629,
|
|
"mean_token_accuracy": 0.9105051755905151,
|
|
"step": 143
|
|
},
|
|
{
|
|
"epoch": 0.6697674418604651,
|
|
"grad_norm": 2.4110724925994873,
|
|
"learning_rate": 3.75257912766578e-06,
|
|
"loss": 0.3487,
|
|
"mean_token_accuracy": 0.8845166563987732,
|
|
"step": 144
|
|
},
|
|
{
|
|
"epoch": 0.6744186046511628,
|
|
"grad_norm": 2.4061532020568848,
|
|
"learning_rate": 3.6853121931559334e-06,
|
|
"loss": 0.3179,
|
|
"mean_token_accuracy": 0.8907623291015625,
|
|
"step": 145
|
|
},
|
|
{
|
|
"epoch": 0.6790697674418604,
|
|
"grad_norm": 2.232574701309204,
|
|
"learning_rate": 3.618526073038574e-06,
|
|
"loss": 0.3334,
|
|
"mean_token_accuracy": 0.8821178674697876,
|
|
"step": 146
|
|
},
|
|
{
|
|
"epoch": 0.6837209302325581,
|
|
"grad_norm": 2.342412233352661,
|
|
"learning_rate": 3.552238462769446e-06,
|
|
"loss": 0.2785,
|
|
"mean_token_accuracy": 0.8974854946136475,
|
|
"step": 147
|
|
},
|
|
{
|
|
"epoch": 0.6883720930232559,
|
|
"grad_norm": 2.459444999694824,
|
|
"learning_rate": 3.4864669257205745e-06,
|
|
"loss": 0.2687,
|
|
"mean_token_accuracy": 0.9067319631576538,
|
|
"step": 148
|
|
},
|
|
{
|
|
"epoch": 0.6930232558139535,
|
|
"grad_norm": 2.3122663497924805,
|
|
"learning_rate": 3.4212288885267246e-06,
|
|
"loss": 0.3143,
|
|
"mean_token_accuracy": 0.8928021788597107,
|
|
"step": 149
|
|
},
|
|
{
|
|
"epoch": 0.6976744186046512,
|
|
"grad_norm": 2.461893081665039,
|
|
"learning_rate": 3.3565416364681016e-06,
|
|
"loss": 0.2822,
|
|
"mean_token_accuracy": 0.897793710231781,
|
|
"step": 150
|
|
},
|
|
{
|
|
"epoch": 0.7023255813953488,
|
|
"grad_norm": 2.0806939601898193,
|
|
"learning_rate": 3.2924223088904816e-06,
|
|
"loss": 0.2732,
|
|
"mean_token_accuracy": 0.9053335785865784,
|
|
"step": 151
|
|
},
|
|
{
|
|
"epoch": 0.7069767441860465,
|
|
"grad_norm": 2.5477609634399414,
|
|
"learning_rate": 3.228887894664029e-06,
|
|
"loss": 0.3307,
|
|
"mean_token_accuracy": 0.888381838798523,
|
|
"step": 152
|
|
},
|
|
{
|
|
"epoch": 0.7116279069767442,
|
|
"grad_norm": 2.393155574798584,
|
|
"learning_rate": 3.1659552276819693e-06,
|
|
"loss": 0.3073,
|
|
"mean_token_accuracy": 0.891734778881073,
|
|
"step": 153
|
|
},
|
|
{
|
|
"epoch": 0.7162790697674418,
|
|
"grad_norm": 2.217618942260742,
|
|
"learning_rate": 3.1036409824003324e-06,
|
|
"loss": 0.2788,
|
|
"mean_token_accuracy": 0.9050260782241821,
|
|
"step": 154
|
|
},
|
|
{
|
|
"epoch": 0.7209302325581395,
|
|
"grad_norm": 2.061199903488159,
|
|
"learning_rate": 3.0419616694199327e-06,
|
|
"loss": 0.2581,
|
|
"mean_token_accuracy": 0.9108189940452576,
|
|
"step": 155
|
|
},
|
|
{
|
|
"epoch": 0.7255813953488373,
|
|
"grad_norm": 2.1402981281280518,
|
|
"learning_rate": 2.98093363111178e-06,
|
|
"loss": 0.2353,
|
|
"mean_token_accuracy": 0.916283905506134,
|
|
"step": 156
|
|
},
|
|
{
|
|
"epoch": 0.7302325581395349,
|
|
"grad_norm": 2.556551456451416,
|
|
"learning_rate": 2.92057303728704e-06,
|
|
"loss": 0.2929,
|
|
"mean_token_accuracy": 0.8986952900886536,
|
|
"step": 157
|
|
},
|
|
{
|
|
"epoch": 0.7348837209302326,
|
|
"grad_norm": 2.191469669342041,
|
|
"learning_rate": 2.860895880912735e-06,
|
|
"loss": 0.2787,
|
|
"mean_token_accuracy": 0.9077619314193726,
|
|
"step": 158
|
|
},
|
|
{
|
|
"epoch": 0.7395348837209302,
|
|
"grad_norm": 2.2681868076324463,
|
|
"learning_rate": 2.801917973874294e-06,
|
|
"loss": 0.2982,
|
|
"mean_token_accuracy": 0.8943151831626892,
|
|
"step": 159
|
|
},
|
|
{
|
|
"epoch": 0.7441860465116279,
|
|
"grad_norm": 2.360569477081299,
|
|
"learning_rate": 2.7436549427860766e-06,
|
|
"loss": 0.3169,
|
|
"mean_token_accuracy": 0.8914275765419006,
|
|
"step": 160
|
|
},
|
|
{
|
|
"epoch": 0.7488372093023256,
|
|
"grad_norm": 2.0360801219940186,
|
|
"learning_rate": 2.6861222248509926e-06,
|
|
"loss": 0.2436,
|
|
"mean_token_accuracy": 0.9144623875617981,
|
|
"step": 161
|
|
},
|
|
{
|
|
"epoch": 0.7534883720930232,
|
|
"grad_norm": 2.0906593799591064,
|
|
"learning_rate": 2.6293350637703123e-06,
|
|
"loss": 0.2749,
|
|
"mean_token_accuracy": 0.9103024005889893,
|
|
"step": 162
|
|
},
|
|
{
|
|
"epoch": 0.7581395348837209,
|
|
"grad_norm": 2.402468204498291,
|
|
"learning_rate": 2.5733085057047325e-06,
|
|
"loss": 0.2853,
|
|
"mean_token_accuracy": 0.9021291732788086,
|
|
"step": 163
|
|
},
|
|
{
|
|
"epoch": 0.7627906976744186,
|
|
"grad_norm": 2.1489808559417725,
|
|
"learning_rate": 2.518057395287792e-06,
|
|
"loss": 0.2955,
|
|
"mean_token_accuracy": 0.8964577913284302,
|
|
"step": 164
|
|
},
|
|
{
|
|
"epoch": 0.7674418604651163,
|
|
"grad_norm": 2.3413703441619873,
|
|
"learning_rate": 2.463596371692681e-06,
|
|
"loss": 0.2944,
|
|
"mean_token_accuracy": 0.8973432183265686,
|
|
"step": 165
|
|
},
|
|
{
|
|
"epoch": 0.772093023255814,
|
|
"grad_norm": 2.662538528442383,
|
|
"learning_rate": 2.409939864753487e-06,
|
|
"loss": 0.2916,
|
|
"mean_token_accuracy": 0.8977203369140625,
|
|
"step": 166
|
|
},
|
|
{
|
|
"epoch": 0.7767441860465116,
|
|
"grad_norm": 2.1745879650115967,
|
|
"learning_rate": 2.3571020911419067e-06,
|
|
"loss": 0.2797,
|
|
"mean_token_accuracy": 0.9074603319168091,
|
|
"step": 167
|
|
},
|
|
{
|
|
"epoch": 0.7813953488372093,
|
|
"grad_norm": 2.391507387161255,
|
|
"learning_rate": 2.3050970506004463e-06,
|
|
"loss": 0.3075,
|
|
"mean_token_accuracy": 0.893263041973114,
|
|
"step": 168
|
|
},
|
|
{
|
|
"epoch": 0.786046511627907,
|
|
"grad_norm": 2.3911025524139404,
|
|
"learning_rate": 2.2539385222330797e-06,
|
|
"loss": 0.2507,
|
|
"mean_token_accuracy": 0.9109898209571838,
|
|
"step": 169
|
|
},
|
|
{
|
|
"epoch": 0.7906976744186046,
|
|
"grad_norm": 2.23653507232666,
|
|
"learning_rate": 2.203640060854387e-06,
|
|
"loss": 0.2621,
|
|
"mean_token_accuracy": 0.9102239608764648,
|
|
"step": 170
|
|
},
|
|
{
|
|
"epoch": 0.7953488372093023,
|
|
"grad_norm": 2.166534662246704,
|
|
"learning_rate": 2.1542149933981014e-06,
|
|
"loss": 0.2956,
|
|
"mean_token_accuracy": 0.8992091417312622,
|
|
"step": 171
|
|
},
|
|
{
|
|
"epoch": 0.8,
|
|
"grad_norm": 2.120082378387451,
|
|
"learning_rate": 2.10567641538605e-06,
|
|
"loss": 0.2717,
|
|
"mean_token_accuracy": 0.9100556969642639,
|
|
"step": 172
|
|
},
|
|
{
|
|
"epoch": 0.8046511627906977,
|
|
"grad_norm": 2.098536729812622,
|
|
"learning_rate": 2.058037187458398e-06,
|
|
"loss": 0.2844,
|
|
"mean_token_accuracy": 0.9054663777351379,
|
|
"step": 173
|
|
},
|
|
{
|
|
"epoch": 0.8093023255813954,
|
|
"grad_norm": 2.382420063018799,
|
|
"learning_rate": 2.011309931966136e-06,
|
|
"loss": 0.2783,
|
|
"mean_token_accuracy": 0.9038222432136536,
|
|
"step": 174
|
|
},
|
|
{
|
|
"epoch": 0.813953488372093,
|
|
"grad_norm": 2.162177085876465,
|
|
"learning_rate": 1.965507029626695e-06,
|
|
"loss": 0.2869,
|
|
"mean_token_accuracy": 0.8897825479507446,
|
|
"step": 175
|
|
},
|
|
{
|
|
"epoch": 0.8186046511627907,
|
|
"grad_norm": 2.17085337638855,
|
|
"learning_rate": 1.920640616243589e-06,
|
|
"loss": 0.2558,
|
|
"mean_token_accuracy": 0.911536455154419,
|
|
"step": 176
|
|
},
|
|
{
|
|
"epoch": 0.8232558139534883,
|
|
"grad_norm": 2.4038584232330322,
|
|
"learning_rate": 1.8767225794909484e-06,
|
|
"loss": 0.3106,
|
|
"mean_token_accuracy": 0.8924676179885864,
|
|
"step": 177
|
|
},
|
|
{
|
|
"epoch": 0.827906976744186,
|
|
"grad_norm": 2.1180429458618164,
|
|
"learning_rate": 1.8337645557637929e-06,
|
|
"loss": 0.2698,
|
|
"mean_token_accuracy": 0.904086172580719,
|
|
"step": 178
|
|
},
|
|
{
|
|
"epoch": 0.8325581395348837,
|
|
"grad_norm": 2.0960586071014404,
|
|
"learning_rate": 1.7917779270948887e-06,
|
|
"loss": 0.2869,
|
|
"mean_token_accuracy": 0.8998658657073975,
|
|
"step": 179
|
|
},
|
|
{
|
|
"epoch": 0.8372093023255814,
|
|
"grad_norm": 2.246081829071045,
|
|
"learning_rate": 1.7507738181390027e-06,
|
|
"loss": 0.2747,
|
|
"mean_token_accuracy": 0.9040079116821289,
|
|
"step": 180
|
|
},
|
|
{
|
|
"epoch": 0.8418604651162791,
|
|
"grad_norm": 2.2984302043914795,
|
|
"learning_rate": 1.7107630932253383e-06,
|
|
"loss": 0.2802,
|
|
"mean_token_accuracy": 0.9013429880142212,
|
|
"step": 181
|
|
},
|
|
{
|
|
"epoch": 0.8465116279069768,
|
|
"grad_norm": 2.207886219024658,
|
|
"learning_rate": 1.6717563534789594e-06,
|
|
"loss": 0.2851,
|
|
"mean_token_accuracy": 0.9047191739082336,
|
|
"step": 182
|
|
},
|
|
{
|
|
"epoch": 0.8511627906976744,
|
|
"grad_norm": 2.2271370887756348,
|
|
"learning_rate": 1.6337639340119476e-06,
|
|
"loss": 0.2892,
|
|
"mean_token_accuracy": 0.8965415954589844,
|
|
"step": 183
|
|
},
|
|
{
|
|
"epoch": 0.8558139534883721,
|
|
"grad_norm": 2.3133997917175293,
|
|
"learning_rate": 1.596795901185037e-06,
|
|
"loss": 0.2976,
|
|
"mean_token_accuracy": 0.8997429013252258,
|
|
"step": 184
|
|
},
|
|
{
|
|
"epoch": 0.8604651162790697,
|
|
"grad_norm": 1.9818381071090698,
|
|
"learning_rate": 1.5608620499404628e-06,
|
|
"loss": 0.2686,
|
|
"mean_token_accuracy": 0.9057079553604126,
|
|
"step": 185
|
|
},
|
|
{
|
|
"epoch": 0.8651162790697674,
|
|
"grad_norm": 2.238173246383667,
|
|
"learning_rate": 1.5259719012067249e-06,
|
|
"loss": 0.2986,
|
|
"mean_token_accuracy": 0.8974149227142334,
|
|
"step": 186
|
|
},
|
|
{
|
|
"epoch": 0.8697674418604651,
|
|
"grad_norm": 2.2305054664611816,
|
|
"learning_rate": 1.4921346993759453e-06,
|
|
"loss": 0.262,
|
|
"mean_token_accuracy": 0.9045342206954956,
|
|
"step": 187
|
|
},
|
|
{
|
|
"epoch": 0.8744186046511628,
|
|
"grad_norm": 2.110743761062622,
|
|
"learning_rate": 1.459359409854505e-06,
|
|
"loss": 0.2784,
|
|
"mean_token_accuracy": 0.9074790477752686,
|
|
"step": 188
|
|
},
|
|
{
|
|
"epoch": 0.8790697674418605,
|
|
"grad_norm": 2.1508686542510986,
|
|
"learning_rate": 1.4276547166875946e-06,
|
|
"loss": 0.2569,
|
|
"mean_token_accuracy": 0.9104231595993042,
|
|
"step": 189
|
|
},
|
|
{
|
|
"epoch": 0.8837209302325582,
|
|
"grad_norm": 2.3325693607330322,
|
|
"learning_rate": 1.397029020258313e-06,
|
|
"loss": 0.2764,
|
|
"mean_token_accuracy": 0.8998494148254395,
|
|
"step": 190
|
|
},
|
|
{
|
|
"epoch": 0.8883720930232558,
|
|
"grad_norm": 2.333432912826538,
|
|
"learning_rate": 1.367490435061928e-06,
|
|
"loss": 0.3218,
|
|
"mean_token_accuracy": 0.8857652544975281,
|
|
"step": 191
|
|
},
|
|
{
|
|
"epoch": 0.8930232558139535,
|
|
"grad_norm": 2.131383180618286,
|
|
"learning_rate": 1.3390467875558855e-06,
|
|
"loss": 0.2418,
|
|
"mean_token_accuracy": 0.9137630462646484,
|
|
"step": 192
|
|
},
|
|
{
|
|
"epoch": 0.8976744186046511,
|
|
"grad_norm": 2.1150314807891846,
|
|
"learning_rate": 1.3117056140861317e-06,
|
|
"loss": 0.2811,
|
|
"mean_token_accuracy": 0.9052088260650635,
|
|
"step": 193
|
|
},
|
|
{
|
|
"epoch": 0.9023255813953488,
|
|
"grad_norm": 2.3999621868133545,
|
|
"learning_rate": 1.285474158890304e-06,
|
|
"loss": 0.2552,
|
|
"mean_token_accuracy": 0.9161849617958069,
|
|
"step": 194
|
|
},
|
|
{
|
|
"epoch": 0.9069767441860465,
|
|
"grad_norm": 2.0384693145751953,
|
|
"learning_rate": 1.2603593721783219e-06,
|
|
"loss": 0.2594,
|
|
"mean_token_accuracy": 0.9125847816467285,
|
|
"step": 195
|
|
},
|
|
{
|
|
"epoch": 0.9116279069767442,
|
|
"grad_norm": 2.1255104541778564,
|
|
"learning_rate": 1.2363679082908766e-06,
|
|
"loss": 0.291,
|
|
"mean_token_accuracy": 0.9000939726829529,
|
|
"step": 196
|
|
},
|
|
{
|
|
"epoch": 0.9162790697674419,
|
|
"grad_norm": 2.2071197032928467,
|
|
"learning_rate": 1.2135061239363161e-06,
|
|
"loss": 0.25,
|
|
"mean_token_accuracy": 0.9136289954185486,
|
|
"step": 197
|
|
},
|
|
{
|
|
"epoch": 0.9209302325581395,
|
|
"grad_norm": 2.2119243144989014,
|
|
"learning_rate": 1.1917800765063954e-06,
|
|
"loss": 0.2607,
|
|
"mean_token_accuracy": 0.9083563089370728,
|
|
"step": 198
|
|
},
|
|
{
|
|
"epoch": 0.9255813953488372,
|
|
"grad_norm": 2.6126599311828613,
|
|
"learning_rate": 1.1711955224713209e-06,
|
|
"loss": 0.2884,
|
|
"mean_token_accuracy": 0.9001862406730652,
|
|
"step": 199
|
|
},
|
|
{
|
|
"epoch": 0.9302325581395349,
|
|
"grad_norm": 2.1286816596984863,
|
|
"learning_rate": 1.1517579158545386e-06,
|
|
"loss": 0.2605,
|
|
"mean_token_accuracy": 0.9096025824546814,
|
|
"step": 200
|
|
},
|
|
{
|
|
"epoch": 0.9348837209302325,
|
|
"grad_norm": 2.335995674133301,
|
|
"learning_rate": 1.1334724067876463e-06,
|
|
"loss": 0.2909,
|
|
"mean_token_accuracy": 0.9009100198745728,
|
|
"step": 201
|
|
},
|
|
{
|
|
"epoch": 0.9395348837209302,
|
|
"grad_norm": 2.084406614303589,
|
|
"learning_rate": 1.1163438401458358e-06,
|
|
"loss": 0.2571,
|
|
"mean_token_accuracy": 0.9103472232818604,
|
|
"step": 202
|
|
},
|
|
{
|
|
"epoch": 0.9441860465116279,
|
|
"grad_norm": 2.2988972663879395,
|
|
"learning_rate": 1.1003767542642021e-06,
|
|
"loss": 0.2759,
|
|
"mean_token_accuracy": 0.9071067571640015,
|
|
"step": 203
|
|
},
|
|
{
|
|
"epoch": 0.9488372093023256,
|
|
"grad_norm": 2.0890564918518066,
|
|
"learning_rate": 1.0855753797352868e-06,
|
|
"loss": 0.2423,
|
|
"mean_token_accuracy": 0.9124136567115784,
|
|
"step": 204
|
|
},
|
|
{
|
|
"epoch": 0.9534883720930233,
|
|
"grad_norm": 2.03589129447937,
|
|
"learning_rate": 1.0719436382881466e-06,
|
|
"loss": 0.253,
|
|
"mean_token_accuracy": 0.910382866859436,
|
|
"step": 205
|
|
},
|
|
{
|
|
"epoch": 0.958139534883721,
|
|
"grad_norm": 2.1397945880889893,
|
|
"learning_rate": 1.0594851417492665e-06,
|
|
"loss": 0.2729,
|
|
"mean_token_accuracy": 0.9032517671585083,
|
|
"step": 206
|
|
},
|
|
{
|
|
"epoch": 0.9627906976744186,
|
|
"grad_norm": 2.3552486896514893,
|
|
"learning_rate": 1.0482031910855804e-06,
|
|
"loss": 0.2933,
|
|
"mean_token_accuracy": 0.8961015343666077,
|
|
"step": 207
|
|
},
|
|
{
|
|
"epoch": 0.9674418604651163,
|
|
"grad_norm": 2.2049098014831543,
|
|
"learning_rate": 1.0381007755298547e-06,
|
|
"loss": 0.2682,
|
|
"mean_token_accuracy": 0.9107953310012817,
|
|
"step": 208
|
|
},
|
|
{
|
|
"epoch": 0.9720930232558139,
|
|
"grad_norm": 2.134488105773926,
|
|
"learning_rate": 1.029180571788672e-06,
|
|
"loss": 0.2483,
|
|
"mean_token_accuracy": 0.9142372608184814,
|
|
"step": 209
|
|
},
|
|
{
|
|
"epoch": 0.9767441860465116,
|
|
"grad_norm": 2.0080952644348145,
|
|
"learning_rate": 1.021444943333218e-06,
|
|
"loss": 0.2562,
|
|
"mean_token_accuracy": 0.9109402298927307,
|
|
"step": 210
|
|
},
|
|
{
|
|
"epoch": 0.9813953488372092,
|
|
"grad_norm": 2.1564061641693115,
|
|
"learning_rate": 1.0148959397730637e-06,
|
|
"loss": 0.2525,
|
|
"mean_token_accuracy": 0.9117802381515503,
|
|
"step": 211
|
|
},
|
|
{
|
|
"epoch": 0.986046511627907,
|
|
"grad_norm": 2.045999050140381,
|
|
"learning_rate": 1.0095352963131057e-06,
|
|
"loss": 0.2763,
|
|
"mean_token_accuracy": 0.9055154323577881,
|
|
"step": 212
|
|
},
|
|
{
|
|
"epoch": 0.9906976744186047,
|
|
"grad_norm": 2.1734366416931152,
|
|
"learning_rate": 1.0053644332938118e-06,
|
|
"loss": 0.2456,
|
|
"mean_token_accuracy": 0.9108783602714539,
|
|
"step": 213
|
|
},
|
|
{
|
|
"epoch": 0.9953488372093023,
|
|
"grad_norm": 2.192584753036499,
|
|
"learning_rate": 1.0023844558148912e-06,
|
|
"loss": 0.2888,
|
|
"mean_token_accuracy": 0.9005217552185059,
|
|
"step": 214
|
|
},
|
|
{
|
|
"epoch": 1.0,
|
|
"grad_norm": 1.960463523864746,
|
|
"learning_rate": 1.0005961534424925e-06,
|
|
"loss": 0.2204,
|
|
"mean_token_accuracy": 0.9134988188743591,
|
|
"step": 215
|
|
},
|
|
{
|
|
"epoch": 1.0,
|
|
"step": 215,
|
|
"total_flos": 2.0144468407196058e+17,
|
|
"train_loss": 0.3228219868138779,
|
|
"train_runtime": 1246.4994,
|
|
"train_samples_per_second": 5.496,
|
|
"train_steps_per_second": 0.172
|
|
}
|
|
],
|
|
"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
|
|
}
|