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": 19.56722068786621,
|
|
"learning_rate": 0.0,
|
|
"loss": 0.842,
|
|
"mean_token_accuracy": 0.8234314918518066,
|
|
"step": 1
|
|
},
|
|
{
|
|
"epoch": 0.009302325581395349,
|
|
"grad_norm": 21.499818801879883,
|
|
"learning_rate": 4.5454545454545457e-07,
|
|
"loss": 0.8773,
|
|
"mean_token_accuracy": 0.8220880627632141,
|
|
"step": 2
|
|
},
|
|
{
|
|
"epoch": 0.013953488372093023,
|
|
"grad_norm": 19.097368240356445,
|
|
"learning_rate": 9.090909090909091e-07,
|
|
"loss": 0.8511,
|
|
"mean_token_accuracy": 0.8329019546508789,
|
|
"step": 3
|
|
},
|
|
{
|
|
"epoch": 0.018604651162790697,
|
|
"grad_norm": 18.014055252075195,
|
|
"learning_rate": 1.3636363636363636e-06,
|
|
"loss": 0.8172,
|
|
"mean_token_accuracy": 0.8300436735153198,
|
|
"step": 4
|
|
},
|
|
{
|
|
"epoch": 0.023255813953488372,
|
|
"grad_norm": 21.1191349029541,
|
|
"learning_rate": 1.8181818181818183e-06,
|
|
"loss": 0.932,
|
|
"mean_token_accuracy": 0.8103048801422119,
|
|
"step": 5
|
|
},
|
|
{
|
|
"epoch": 0.027906976744186046,
|
|
"grad_norm": 18.354591369628906,
|
|
"learning_rate": 2.2727272727272728e-06,
|
|
"loss": 0.782,
|
|
"mean_token_accuracy": 0.8292044401168823,
|
|
"step": 6
|
|
},
|
|
{
|
|
"epoch": 0.03255813953488372,
|
|
"grad_norm": 16.53504753112793,
|
|
"learning_rate": 2.7272727272727272e-06,
|
|
"loss": 0.8115,
|
|
"mean_token_accuracy": 0.8264797925949097,
|
|
"step": 7
|
|
},
|
|
{
|
|
"epoch": 0.037209302325581395,
|
|
"grad_norm": 10.243330955505371,
|
|
"learning_rate": 3.181818181818182e-06,
|
|
"loss": 0.6383,
|
|
"mean_token_accuracy": 0.8500951528549194,
|
|
"step": 8
|
|
},
|
|
{
|
|
"epoch": 0.04186046511627907,
|
|
"grad_norm": 8.570775985717773,
|
|
"learning_rate": 3.6363636363636366e-06,
|
|
"loss": 0.5647,
|
|
"mean_token_accuracy": 0.8617518544197083,
|
|
"step": 9
|
|
},
|
|
{
|
|
"epoch": 0.046511627906976744,
|
|
"grad_norm": 9.014289855957031,
|
|
"learning_rate": 4.0909090909090915e-06,
|
|
"loss": 0.4928,
|
|
"mean_token_accuracy": 0.8720639944076538,
|
|
"step": 10
|
|
},
|
|
{
|
|
"epoch": 0.05116279069767442,
|
|
"grad_norm": 9.353399276733398,
|
|
"learning_rate": 4.5454545454545455e-06,
|
|
"loss": 0.4215,
|
|
"mean_token_accuracy": 0.8906779885292053,
|
|
"step": 11
|
|
},
|
|
{
|
|
"epoch": 0.05581395348837209,
|
|
"grad_norm": 7.49207067489624,
|
|
"learning_rate": 5e-06,
|
|
"loss": 0.4415,
|
|
"mean_token_accuracy": 0.8809018135070801,
|
|
"step": 12
|
|
},
|
|
{
|
|
"epoch": 0.06046511627906977,
|
|
"grad_norm": 4.262305736541748,
|
|
"learning_rate": 5.4545454545454545e-06,
|
|
"loss": 0.42,
|
|
"mean_token_accuracy": 0.8770877122879028,
|
|
"step": 13
|
|
},
|
|
{
|
|
"epoch": 0.06511627906976744,
|
|
"grad_norm": 5.891422748565674,
|
|
"learning_rate": 5.90909090909091e-06,
|
|
"loss": 0.3657,
|
|
"mean_token_accuracy": 0.8887879252433777,
|
|
"step": 14
|
|
},
|
|
{
|
|
"epoch": 0.06976744186046512,
|
|
"grad_norm": 3.392139196395874,
|
|
"learning_rate": 6.363636363636364e-06,
|
|
"loss": 0.3438,
|
|
"mean_token_accuracy": 0.8969727158546448,
|
|
"step": 15
|
|
},
|
|
{
|
|
"epoch": 0.07441860465116279,
|
|
"grad_norm": 3.7728092670440674,
|
|
"learning_rate": 6.818181818181818e-06,
|
|
"loss": 0.381,
|
|
"mean_token_accuracy": 0.8863928318023682,
|
|
"step": 16
|
|
},
|
|
{
|
|
"epoch": 0.07906976744186046,
|
|
"grad_norm": 3.083871603012085,
|
|
"learning_rate": 7.272727272727273e-06,
|
|
"loss": 0.333,
|
|
"mean_token_accuracy": 0.902701735496521,
|
|
"step": 17
|
|
},
|
|
{
|
|
"epoch": 0.08372093023255814,
|
|
"grad_norm": 3.078378200531006,
|
|
"learning_rate": 7.727272727272727e-06,
|
|
"loss": 0.3776,
|
|
"mean_token_accuracy": 0.8830382823944092,
|
|
"step": 18
|
|
},
|
|
{
|
|
"epoch": 0.08837209302325581,
|
|
"grad_norm": 3.3377773761749268,
|
|
"learning_rate": 8.181818181818183e-06,
|
|
"loss": 0.384,
|
|
"mean_token_accuracy": 0.8844019174575806,
|
|
"step": 19
|
|
},
|
|
{
|
|
"epoch": 0.09302325581395349,
|
|
"grad_norm": 2.8573648929595947,
|
|
"learning_rate": 8.636363636363637e-06,
|
|
"loss": 0.3525,
|
|
"mean_token_accuracy": 0.8916018605232239,
|
|
"step": 20
|
|
},
|
|
{
|
|
"epoch": 0.09767441860465116,
|
|
"grad_norm": 3.159400463104248,
|
|
"learning_rate": 9.090909090909091e-06,
|
|
"loss": 0.3996,
|
|
"mean_token_accuracy": 0.8746355772018433,
|
|
"step": 21
|
|
},
|
|
{
|
|
"epoch": 0.10232558139534884,
|
|
"grad_norm": 3.09920334815979,
|
|
"learning_rate": 9.545454545454547e-06,
|
|
"loss": 0.3518,
|
|
"mean_token_accuracy": 0.8835098147392273,
|
|
"step": 22
|
|
},
|
|
{
|
|
"epoch": 0.10697674418604651,
|
|
"grad_norm": 2.9805715084075928,
|
|
"learning_rate": 1e-05,
|
|
"loss": 0.3449,
|
|
"mean_token_accuracy": 0.8943809270858765,
|
|
"step": 23
|
|
},
|
|
{
|
|
"epoch": 0.11162790697674418,
|
|
"grad_norm": 3.226511001586914,
|
|
"learning_rate": 9.999403846557509e-06,
|
|
"loss": 0.3707,
|
|
"mean_token_accuracy": 0.8781068921089172,
|
|
"step": 24
|
|
},
|
|
{
|
|
"epoch": 0.11627906976744186,
|
|
"grad_norm": 3.0211565494537354,
|
|
"learning_rate": 9.99761554418511e-06,
|
|
"loss": 0.3333,
|
|
"mean_token_accuracy": 0.8967213034629822,
|
|
"step": 25
|
|
},
|
|
{
|
|
"epoch": 0.12093023255813953,
|
|
"grad_norm": 3.0328941345214844,
|
|
"learning_rate": 9.99463556670619e-06,
|
|
"loss": 0.3337,
|
|
"mean_token_accuracy": 0.9013436436653137,
|
|
"step": 26
|
|
},
|
|
{
|
|
"epoch": 0.12558139534883722,
|
|
"grad_norm": 3.127363681793213,
|
|
"learning_rate": 9.990464703686895e-06,
|
|
"loss": 0.3643,
|
|
"mean_token_accuracy": 0.8956908583641052,
|
|
"step": 27
|
|
},
|
|
{
|
|
"epoch": 0.13023255813953488,
|
|
"grad_norm": 3.1299784183502197,
|
|
"learning_rate": 9.985104060226937e-06,
|
|
"loss": 0.3536,
|
|
"mean_token_accuracy": 0.8928705453872681,
|
|
"step": 28
|
|
},
|
|
{
|
|
"epoch": 0.13488372093023257,
|
|
"grad_norm": 3.025752305984497,
|
|
"learning_rate": 9.978555056666784e-06,
|
|
"loss": 0.3768,
|
|
"mean_token_accuracy": 0.8789021372795105,
|
|
"step": 29
|
|
},
|
|
{
|
|
"epoch": 0.13953488372093023,
|
|
"grad_norm": 3.072669744491577,
|
|
"learning_rate": 9.97081942821133e-06,
|
|
"loss": 0.3708,
|
|
"mean_token_accuracy": 0.8879210948944092,
|
|
"step": 30
|
|
},
|
|
{
|
|
"epoch": 0.14418604651162792,
|
|
"grad_norm": 3.2670693397521973,
|
|
"learning_rate": 9.961899224470146e-06,
|
|
"loss": 0.3947,
|
|
"mean_token_accuracy": 0.8810369372367859,
|
|
"step": 31
|
|
},
|
|
{
|
|
"epoch": 0.14883720930232558,
|
|
"grad_norm": 3.0507073402404785,
|
|
"learning_rate": 9.95179680891442e-06,
|
|
"loss": 0.3177,
|
|
"mean_token_accuracy": 0.9065743684768677,
|
|
"step": 32
|
|
},
|
|
{
|
|
"epoch": 0.15348837209302327,
|
|
"grad_norm": 3.6491124629974365,
|
|
"learning_rate": 9.940514858250736e-06,
|
|
"loss": 0.3565,
|
|
"mean_token_accuracy": 0.8926541209220886,
|
|
"step": 33
|
|
},
|
|
{
|
|
"epoch": 0.15813953488372093,
|
|
"grad_norm": 2.94435977935791,
|
|
"learning_rate": 9.928056361711854e-06,
|
|
"loss": 0.3032,
|
|
"mean_token_accuracy": 0.9051738977432251,
|
|
"step": 34
|
|
},
|
|
{
|
|
"epoch": 0.16279069767441862,
|
|
"grad_norm": 3.0378453731536865,
|
|
"learning_rate": 9.914424620264714e-06,
|
|
"loss": 0.3582,
|
|
"mean_token_accuracy": 0.8896961808204651,
|
|
"step": 35
|
|
},
|
|
{
|
|
"epoch": 0.16744186046511628,
|
|
"grad_norm": 3.0237748622894287,
|
|
"learning_rate": 9.899623245735798e-06,
|
|
"loss": 0.3287,
|
|
"mean_token_accuracy": 0.8992047309875488,
|
|
"step": 36
|
|
},
|
|
{
|
|
"epoch": 0.17209302325581396,
|
|
"grad_norm": 3.0034892559051514,
|
|
"learning_rate": 9.883656159854166e-06,
|
|
"loss": 0.3175,
|
|
"mean_token_accuracy": 0.9017576575279236,
|
|
"step": 37
|
|
},
|
|
{
|
|
"epoch": 0.17674418604651163,
|
|
"grad_norm": 3.0133862495422363,
|
|
"learning_rate": 9.866527593212355e-06,
|
|
"loss": 0.3321,
|
|
"mean_token_accuracy": 0.8984139561653137,
|
|
"step": 38
|
|
},
|
|
{
|
|
"epoch": 0.1813953488372093,
|
|
"grad_norm": 2.9698572158813477,
|
|
"learning_rate": 9.848242084145462e-06,
|
|
"loss": 0.3203,
|
|
"mean_token_accuracy": 0.9006622433662415,
|
|
"step": 39
|
|
},
|
|
{
|
|
"epoch": 0.18604651162790697,
|
|
"grad_norm": 2.8872194290161133,
|
|
"learning_rate": 9.82880447752868e-06,
|
|
"loss": 0.3377,
|
|
"mean_token_accuracy": 0.8935135006904602,
|
|
"step": 40
|
|
},
|
|
{
|
|
"epoch": 0.19069767441860466,
|
|
"grad_norm": 3.0155012607574463,
|
|
"learning_rate": 9.808219923493606e-06,
|
|
"loss": 0.3318,
|
|
"mean_token_accuracy": 0.8975328803062439,
|
|
"step": 41
|
|
},
|
|
{
|
|
"epoch": 0.19534883720930232,
|
|
"grad_norm": 3.193694591522217,
|
|
"learning_rate": 9.786493876063685e-06,
|
|
"loss": 0.3284,
|
|
"mean_token_accuracy": 0.8954758048057556,
|
|
"step": 42
|
|
},
|
|
{
|
|
"epoch": 0.2,
|
|
"grad_norm": 3.0788753032684326,
|
|
"learning_rate": 9.763632091709125e-06,
|
|
"loss": 0.3093,
|
|
"mean_token_accuracy": 0.8967692852020264,
|
|
"step": 43
|
|
},
|
|
{
|
|
"epoch": 0.20465116279069767,
|
|
"grad_norm": 2.8875858783721924,
|
|
"learning_rate": 9.739640627821678e-06,
|
|
"loss": 0.3373,
|
|
"mean_token_accuracy": 0.8984333872795105,
|
|
"step": 44
|
|
},
|
|
{
|
|
"epoch": 0.20930232558139536,
|
|
"grad_norm": 3.2467026710510254,
|
|
"learning_rate": 9.714525841109697e-06,
|
|
"loss": 0.2945,
|
|
"mean_token_accuracy": 0.9045419692993164,
|
|
"step": 45
|
|
},
|
|
{
|
|
"epoch": 0.21395348837209302,
|
|
"grad_norm": 3.0920603275299072,
|
|
"learning_rate": 9.68829438591387e-06,
|
|
"loss": 0.3045,
|
|
"mean_token_accuracy": 0.9002149701118469,
|
|
"step": 46
|
|
},
|
|
{
|
|
"epoch": 0.2186046511627907,
|
|
"grad_norm": 2.996901512145996,
|
|
"learning_rate": 9.660953212444116e-06,
|
|
"loss": 0.3139,
|
|
"mean_token_accuracy": 0.8980564475059509,
|
|
"step": 47
|
|
},
|
|
{
|
|
"epoch": 0.22325581395348837,
|
|
"grad_norm": 2.84829044342041,
|
|
"learning_rate": 9.632509564938073e-06,
|
|
"loss": 0.3218,
|
|
"mean_token_accuracy": 0.8964840173721313,
|
|
"step": 48
|
|
},
|
|
{
|
|
"epoch": 0.22790697674418606,
|
|
"grad_norm": 3.001344680786133,
|
|
"learning_rate": 9.60297097974169e-06,
|
|
"loss": 0.3145,
|
|
"mean_token_accuracy": 0.8978151082992554,
|
|
"step": 49
|
|
},
|
|
{
|
|
"epoch": 0.23255813953488372,
|
|
"grad_norm": 2.866776943206787,
|
|
"learning_rate": 9.572345283312407e-06,
|
|
"loss": 0.3155,
|
|
"mean_token_accuracy": 0.8966426253318787,
|
|
"step": 50
|
|
},
|
|
{
|
|
"epoch": 0.2372093023255814,
|
|
"grad_norm": 3.1876516342163086,
|
|
"learning_rate": 9.540640590145496e-06,
|
|
"loss": 0.3358,
|
|
"mean_token_accuracy": 0.8820154070854187,
|
|
"step": 51
|
|
},
|
|
{
|
|
"epoch": 0.24186046511627907,
|
|
"grad_norm": 2.8794522285461426,
|
|
"learning_rate": 9.507865300624057e-06,
|
|
"loss": 0.3378,
|
|
"mean_token_accuracy": 0.8850215673446655,
|
|
"step": 52
|
|
},
|
|
{
|
|
"epoch": 0.24651162790697675,
|
|
"grad_norm": 2.9620361328125,
|
|
"learning_rate": 9.474028098793277e-06,
|
|
"loss": 0.3129,
|
|
"mean_token_accuracy": 0.8966131806373596,
|
|
"step": 53
|
|
},
|
|
{
|
|
"epoch": 0.25116279069767444,
|
|
"grad_norm": 2.8167107105255127,
|
|
"learning_rate": 9.439137950059539e-06,
|
|
"loss": 0.3294,
|
|
"mean_token_accuracy": 0.8901098966598511,
|
|
"step": 54
|
|
},
|
|
{
|
|
"epoch": 0.2558139534883721,
|
|
"grad_norm": 2.7492761611938477,
|
|
"learning_rate": 9.403204098814965e-06,
|
|
"loss": 0.3288,
|
|
"mean_token_accuracy": 0.8896990418434143,
|
|
"step": 55
|
|
},
|
|
{
|
|
"epoch": 0.26046511627906976,
|
|
"grad_norm": 2.7716281414031982,
|
|
"learning_rate": 9.366236065988053e-06,
|
|
"loss": 0.321,
|
|
"mean_token_accuracy": 0.896246075630188,
|
|
"step": 56
|
|
},
|
|
{
|
|
"epoch": 0.2651162790697674,
|
|
"grad_norm": 2.415069341659546,
|
|
"learning_rate": 9.32824364652104e-06,
|
|
"loss": 0.2824,
|
|
"mean_token_accuracy": 0.9081918597221375,
|
|
"step": 57
|
|
},
|
|
{
|
|
"epoch": 0.26976744186046514,
|
|
"grad_norm": 2.4787442684173584,
|
|
"learning_rate": 9.289236906774663e-06,
|
|
"loss": 0.3112,
|
|
"mean_token_accuracy": 0.8966255187988281,
|
|
"step": 58
|
|
},
|
|
{
|
|
"epoch": 0.2744186046511628,
|
|
"grad_norm": 2.7912585735321045,
|
|
"learning_rate": 9.249226181861e-06,
|
|
"loss": 0.2924,
|
|
"mean_token_accuracy": 0.9057479500770569,
|
|
"step": 59
|
|
},
|
|
{
|
|
"epoch": 0.27906976744186046,
|
|
"grad_norm": 2.6348111629486084,
|
|
"learning_rate": 9.208222072905113e-06,
|
|
"loss": 0.3073,
|
|
"mean_token_accuracy": 0.8927038908004761,
|
|
"step": 60
|
|
},
|
|
{
|
|
"epoch": 0.2837209302325581,
|
|
"grad_norm": 2.7154784202575684,
|
|
"learning_rate": 9.166235444236209e-06,
|
|
"loss": 0.3418,
|
|
"mean_token_accuracy": 0.883296012878418,
|
|
"step": 61
|
|
},
|
|
{
|
|
"epoch": 0.28837209302325584,
|
|
"grad_norm": 2.7209088802337646,
|
|
"learning_rate": 9.123277420509053e-06,
|
|
"loss": 0.3355,
|
|
"mean_token_accuracy": 0.8873890042304993,
|
|
"step": 62
|
|
},
|
|
{
|
|
"epoch": 0.2930232558139535,
|
|
"grad_norm": 2.58394193649292,
|
|
"learning_rate": 9.079359383756411e-06,
|
|
"loss": 0.3137,
|
|
"mean_token_accuracy": 0.8987603187561035,
|
|
"step": 63
|
|
},
|
|
{
|
|
"epoch": 0.29767441860465116,
|
|
"grad_norm": 2.5450754165649414,
|
|
"learning_rate": 9.034492970373305e-06,
|
|
"loss": 0.3055,
|
|
"mean_token_accuracy": 0.8960996270179749,
|
|
"step": 64
|
|
},
|
|
{
|
|
"epoch": 0.3023255813953488,
|
|
"grad_norm": 2.6043131351470947,
|
|
"learning_rate": 8.988690068033864e-06,
|
|
"loss": 0.3329,
|
|
"mean_token_accuracy": 0.8888005018234253,
|
|
"step": 65
|
|
},
|
|
{
|
|
"epoch": 0.30697674418604654,
|
|
"grad_norm": 2.310673475265503,
|
|
"learning_rate": 8.941962812541604e-06,
|
|
"loss": 0.3013,
|
|
"mean_token_accuracy": 0.8942438960075378,
|
|
"step": 66
|
|
},
|
|
{
|
|
"epoch": 0.3116279069767442,
|
|
"grad_norm": 2.518235206604004,
|
|
"learning_rate": 8.894323584613951e-06,
|
|
"loss": 0.3199,
|
|
"mean_token_accuracy": 0.8883762359619141,
|
|
"step": 67
|
|
},
|
|
{
|
|
"epoch": 0.31627906976744186,
|
|
"grad_norm": 2.4000134468078613,
|
|
"learning_rate": 8.845785006601898e-06,
|
|
"loss": 0.2818,
|
|
"mean_token_accuracy": 0.9015190601348877,
|
|
"step": 68
|
|
},
|
|
{
|
|
"epoch": 0.3209302325581395,
|
|
"grad_norm": 2.541795492172241,
|
|
"learning_rate": 8.796359939145614e-06,
|
|
"loss": 0.3087,
|
|
"mean_token_accuracy": 0.8922914266586304,
|
|
"step": 69
|
|
},
|
|
{
|
|
"epoch": 0.32558139534883723,
|
|
"grad_norm": 2.65686297416687,
|
|
"learning_rate": 8.74606147776692e-06,
|
|
"loss": 0.3094,
|
|
"mean_token_accuracy": 0.8930131196975708,
|
|
"step": 70
|
|
},
|
|
{
|
|
"epoch": 0.3302325581395349,
|
|
"grad_norm": 2.518319606781006,
|
|
"learning_rate": 8.694902949399555e-06,
|
|
"loss": 0.3091,
|
|
"mean_token_accuracy": 0.8983761072158813,
|
|
"step": 71
|
|
},
|
|
{
|
|
"epoch": 0.33488372093023255,
|
|
"grad_norm": 2.572589159011841,
|
|
"learning_rate": 8.642897908858096e-06,
|
|
"loss": 0.3108,
|
|
"mean_token_accuracy": 0.8960880041122437,
|
|
"step": 72
|
|
},
|
|
{
|
|
"epoch": 0.3395348837209302,
|
|
"grad_norm": 2.646913766860962,
|
|
"learning_rate": 8.590060135246516e-06,
|
|
"loss": 0.3362,
|
|
"mean_token_accuracy": 0.8827519416809082,
|
|
"step": 73
|
|
},
|
|
{
|
|
"epoch": 0.34418604651162793,
|
|
"grad_norm": 2.309241771697998,
|
|
"learning_rate": 8.53640362830732e-06,
|
|
"loss": 0.2904,
|
|
"mean_token_accuracy": 0.8993450403213501,
|
|
"step": 74
|
|
},
|
|
{
|
|
"epoch": 0.3488372093023256,
|
|
"grad_norm": 2.6384401321411133,
|
|
"learning_rate": 8.481942604712209e-06,
|
|
"loss": 0.3195,
|
|
"mean_token_accuracy": 0.8918577432632446,
|
|
"step": 75
|
|
},
|
|
{
|
|
"epoch": 0.35348837209302325,
|
|
"grad_norm": 2.4883103370666504,
|
|
"learning_rate": 8.426691494295269e-06,
|
|
"loss": 0.2889,
|
|
"mean_token_accuracy": 0.9023594856262207,
|
|
"step": 76
|
|
},
|
|
{
|
|
"epoch": 0.3581395348837209,
|
|
"grad_norm": 2.464266538619995,
|
|
"learning_rate": 8.370664936229688e-06,
|
|
"loss": 0.3144,
|
|
"mean_token_accuracy": 0.8927644491195679,
|
|
"step": 77
|
|
},
|
|
{
|
|
"epoch": 0.3627906976744186,
|
|
"grad_norm": 2.4750561714172363,
|
|
"learning_rate": 8.313877775149009e-06,
|
|
"loss": 0.2742,
|
|
"mean_token_accuracy": 0.9066977500915527,
|
|
"step": 78
|
|
},
|
|
{
|
|
"epoch": 0.3674418604651163,
|
|
"grad_norm": 2.781059980392456,
|
|
"learning_rate": 8.256345057213925e-06,
|
|
"loss": 0.3152,
|
|
"mean_token_accuracy": 0.8888514041900635,
|
|
"step": 79
|
|
},
|
|
{
|
|
"epoch": 0.37209302325581395,
|
|
"grad_norm": 2.3490397930145264,
|
|
"learning_rate": 8.198082026125707e-06,
|
|
"loss": 0.3281,
|
|
"mean_token_accuracy": 0.8870996236801147,
|
|
"step": 80
|
|
},
|
|
{
|
|
"epoch": 0.3767441860465116,
|
|
"grad_norm": 2.4752197265625,
|
|
"learning_rate": 8.139104119087265e-06,
|
|
"loss": 0.2822,
|
|
"mean_token_accuracy": 0.9008604288101196,
|
|
"step": 81
|
|
},
|
|
{
|
|
"epoch": 0.3813953488372093,
|
|
"grad_norm": 2.4564919471740723,
|
|
"learning_rate": 8.07942696271296e-06,
|
|
"loss": 0.2819,
|
|
"mean_token_accuracy": 0.9009503126144409,
|
|
"step": 82
|
|
},
|
|
{
|
|
"epoch": 0.386046511627907,
|
|
"grad_norm": 2.3468544483184814,
|
|
"learning_rate": 8.019066368888222e-06,
|
|
"loss": 0.2881,
|
|
"mean_token_accuracy": 0.9013717770576477,
|
|
"step": 83
|
|
},
|
|
{
|
|
"epoch": 0.39069767441860465,
|
|
"grad_norm": 2.4819157123565674,
|
|
"learning_rate": 7.958038330580067e-06,
|
|
"loss": 0.3321,
|
|
"mean_token_accuracy": 0.8821220993995667,
|
|
"step": 84
|
|
},
|
|
{
|
|
"epoch": 0.3953488372093023,
|
|
"grad_norm": 2.3831310272216797,
|
|
"learning_rate": 7.89635901759967e-06,
|
|
"loss": 0.2994,
|
|
"mean_token_accuracy": 0.8928157687187195,
|
|
"step": 85
|
|
},
|
|
{
|
|
"epoch": 0.4,
|
|
"grad_norm": 2.7160511016845703,
|
|
"learning_rate": 7.834044772318033e-06,
|
|
"loss": 0.3278,
|
|
"mean_token_accuracy": 0.8914129137992859,
|
|
"step": 86
|
|
},
|
|
{
|
|
"epoch": 0.4046511627906977,
|
|
"grad_norm": 2.4505536556243896,
|
|
"learning_rate": 7.77111210533597e-06,
|
|
"loss": 0.3004,
|
|
"mean_token_accuracy": 0.8894042372703552,
|
|
"step": 87
|
|
},
|
|
{
|
|
"epoch": 0.40930232558139534,
|
|
"grad_norm": 3.191389322280884,
|
|
"learning_rate": 7.707577691109519e-06,
|
|
"loss": 0.3545,
|
|
"mean_token_accuracy": 0.8798402547836304,
|
|
"step": 88
|
|
},
|
|
{
|
|
"epoch": 0.413953488372093,
|
|
"grad_norm": 2.280726671218872,
|
|
"learning_rate": 7.6434583635319e-06,
|
|
"loss": 0.3027,
|
|
"mean_token_accuracy": 0.900479793548584,
|
|
"step": 89
|
|
},
|
|
{
|
|
"epoch": 0.4186046511627907,
|
|
"grad_norm": 2.3006751537323,
|
|
"learning_rate": 7.578771111473276e-06,
|
|
"loss": 0.2587,
|
|
"mean_token_accuracy": 0.9121513962745667,
|
|
"step": 90
|
|
},
|
|
{
|
|
"epoch": 0.4232558139534884,
|
|
"grad_norm": 2.3514935970306396,
|
|
"learning_rate": 7.513533074279427e-06,
|
|
"loss": 0.2674,
|
|
"mean_token_accuracy": 0.908033013343811,
|
|
"step": 91
|
|
},
|
|
{
|
|
"epoch": 0.42790697674418604,
|
|
"grad_norm": 2.483891487121582,
|
|
"learning_rate": 7.4477615372305545e-06,
|
|
"loss": 0.3212,
|
|
"mean_token_accuracy": 0.8946645855903625,
|
|
"step": 92
|
|
},
|
|
{
|
|
"epoch": 0.4325581395348837,
|
|
"grad_norm": 2.3761796951293945,
|
|
"learning_rate": 7.3814739269614265e-06,
|
|
"loss": 0.2946,
|
|
"mean_token_accuracy": 0.900073230266571,
|
|
"step": 93
|
|
},
|
|
{
|
|
"epoch": 0.4372093023255814,
|
|
"grad_norm": 2.4102704524993896,
|
|
"learning_rate": 7.314687806844067e-06,
|
|
"loss": 0.3073,
|
|
"mean_token_accuracy": 0.8981302380561829,
|
|
"step": 94
|
|
},
|
|
{
|
|
"epoch": 0.4418604651162791,
|
|
"grad_norm": 2.468144416809082,
|
|
"learning_rate": 7.247420872334221e-06,
|
|
"loss": 0.302,
|
|
"mean_token_accuracy": 0.8967844843864441,
|
|
"step": 95
|
|
},
|
|
{
|
|
"epoch": 0.44651162790697674,
|
|
"grad_norm": 2.222743034362793,
|
|
"learning_rate": 7.179690946282808e-06,
|
|
"loss": 0.2823,
|
|
"mean_token_accuracy": 0.9029505848884583,
|
|
"step": 96
|
|
},
|
|
{
|
|
"epoch": 0.4511627906976744,
|
|
"grad_norm": 2.301776647567749,
|
|
"learning_rate": 7.111515974213639e-06,
|
|
"loss": 0.2494,
|
|
"mean_token_accuracy": 0.9104065895080566,
|
|
"step": 97
|
|
},
|
|
{
|
|
"epoch": 0.4558139534883721,
|
|
"grad_norm": 2.373887777328491,
|
|
"learning_rate": 7.042914019568621e-06,
|
|
"loss": 0.3023,
|
|
"mean_token_accuracy": 0.8933764100074768,
|
|
"step": 98
|
|
},
|
|
{
|
|
"epoch": 0.4604651162790698,
|
|
"grad_norm": 2.228679656982422,
|
|
"learning_rate": 6.973903258921719e-06,
|
|
"loss": 0.2563,
|
|
"mean_token_accuracy": 0.910088062286377,
|
|
"step": 99
|
|
},
|
|
{
|
|
"epoch": 0.46511627906976744,
|
|
"grad_norm": 2.3759074211120605,
|
|
"learning_rate": 6.904501977162949e-06,
|
|
"loss": 0.3274,
|
|
"mean_token_accuracy": 0.8846719861030579,
|
|
"step": 100
|
|
},
|
|
{
|
|
"epoch": 0.4697674418604651,
|
|
"grad_norm": 2.367203712463379,
|
|
"learning_rate": 6.834728562653659e-06,
|
|
"loss": 0.3091,
|
|
"mean_token_accuracy": 0.8904908299446106,
|
|
"step": 101
|
|
},
|
|
{
|
|
"epoch": 0.4744186046511628,
|
|
"grad_norm": 2.4363155364990234,
|
|
"learning_rate": 6.764601502354403e-06,
|
|
"loss": 0.3064,
|
|
"mean_token_accuracy": 0.8976874351501465,
|
|
"step": 102
|
|
},
|
|
{
|
|
"epoch": 0.4790697674418605,
|
|
"grad_norm": 2.2353415489196777,
|
|
"learning_rate": 6.6941393769266995e-06,
|
|
"loss": 0.2975,
|
|
"mean_token_accuracy": 0.8998541831970215,
|
|
"step": 103
|
|
},
|
|
{
|
|
"epoch": 0.48372093023255813,
|
|
"grad_norm": 2.108872890472412,
|
|
"learning_rate": 6.6233608558099405e-06,
|
|
"loss": 0.2778,
|
|
"mean_token_accuracy": 0.9046753644943237,
|
|
"step": 104
|
|
},
|
|
{
|
|
"epoch": 0.4883720930232558,
|
|
"grad_norm": 2.300018310546875,
|
|
"learning_rate": 6.552284692274803e-06,
|
|
"loss": 0.3093,
|
|
"mean_token_accuracy": 0.8958457112312317,
|
|
"step": 105
|
|
},
|
|
{
|
|
"epoch": 0.4930232558139535,
|
|
"grad_norm": 2.3953607082366943,
|
|
"learning_rate": 6.48092971845443e-06,
|
|
"loss": 0.2805,
|
|
"mean_token_accuracy": 0.9026817083358765,
|
|
"step": 106
|
|
},
|
|
{
|
|
"epoch": 0.49767441860465117,
|
|
"grad_norm": 2.397383451461792,
|
|
"learning_rate": 6.409314840354724e-06,
|
|
"loss": 0.3381,
|
|
"mean_token_accuracy": 0.8877024054527283,
|
|
"step": 107
|
|
},
|
|
{
|
|
"epoch": 0.5023255813953489,
|
|
"grad_norm": 2.524808883666992,
|
|
"learning_rate": 6.337459032845068e-06,
|
|
"loss": 0.3136,
|
|
"mean_token_accuracy": 0.8877217173576355,
|
|
"step": 108
|
|
},
|
|
{
|
|
"epoch": 0.5069767441860465,
|
|
"grad_norm": 2.4271292686462402,
|
|
"learning_rate": 6.2653813346308e-06,
|
|
"loss": 0.2741,
|
|
"mean_token_accuracy": 0.9050701260566711,
|
|
"step": 109
|
|
},
|
|
{
|
|
"epoch": 0.5116279069767442,
|
|
"grad_norm": 2.2498927116394043,
|
|
"learning_rate": 6.193100843208772e-06,
|
|
"loss": 0.2809,
|
|
"mean_token_accuracy": 0.8968031406402588,
|
|
"step": 110
|
|
},
|
|
{
|
|
"epoch": 0.5162790697674419,
|
|
"grad_norm": 2.330934524536133,
|
|
"learning_rate": 6.120636709807334e-06,
|
|
"loss": 0.31,
|
|
"mean_token_accuracy": 0.8886767029762268,
|
|
"step": 111
|
|
},
|
|
{
|
|
"epoch": 0.5209302325581395,
|
|
"grad_norm": 2.399275302886963,
|
|
"learning_rate": 6.048008134312078e-06,
|
|
"loss": 0.2888,
|
|
"mean_token_accuracy": 0.9004721641540527,
|
|
"step": 112
|
|
},
|
|
{
|
|
"epoch": 0.5255813953488372,
|
|
"grad_norm": 2.39941668510437,
|
|
"learning_rate": 5.975234360178698e-06,
|
|
"loss": 0.3259,
|
|
"mean_token_accuracy": 0.8926165103912354,
|
|
"step": 113
|
|
},
|
|
{
|
|
"epoch": 0.5302325581395348,
|
|
"grad_norm": 2.22544002532959,
|
|
"learning_rate": 5.902334669334287e-06,
|
|
"loss": 0.2992,
|
|
"mean_token_accuracy": 0.90220046043396,
|
|
"step": 114
|
|
},
|
|
{
|
|
"epoch": 0.5348837209302325,
|
|
"grad_norm": 2.4933297634124756,
|
|
"learning_rate": 5.829328377068476e-06,
|
|
"loss": 0.2943,
|
|
"mean_token_accuracy": 0.8951420783996582,
|
|
"step": 115
|
|
},
|
|
{
|
|
"epoch": 0.5395348837209303,
|
|
"grad_norm": 2.575714349746704,
|
|
"learning_rate": 5.756234826915686e-06,
|
|
"loss": 0.3028,
|
|
"mean_token_accuracy": 0.8980122208595276,
|
|
"step": 116
|
|
},
|
|
{
|
|
"epoch": 0.5441860465116279,
|
|
"grad_norm": 2.6264634132385254,
|
|
"learning_rate": 5.683073385529938e-06,
|
|
"loss": 0.3157,
|
|
"mean_token_accuracy": 0.8895466923713684,
|
|
"step": 117
|
|
},
|
|
{
|
|
"epoch": 0.5488372093023256,
|
|
"grad_norm": 2.182809591293335,
|
|
"learning_rate": 5.60986343755352e-06,
|
|
"loss": 0.2617,
|
|
"mean_token_accuracy": 0.907944917678833,
|
|
"step": 118
|
|
},
|
|
{
|
|
"epoch": 0.5534883720930233,
|
|
"grad_norm": 2.4722118377685547,
|
|
"learning_rate": 5.536624380480878e-06,
|
|
"loss": 0.2513,
|
|
"mean_token_accuracy": 0.9123600125312805,
|
|
"step": 119
|
|
},
|
|
{
|
|
"epoch": 0.5581395348837209,
|
|
"grad_norm": 2.3010671138763428,
|
|
"learning_rate": 5.4633756195191235e-06,
|
|
"loss": 0.2599,
|
|
"mean_token_accuracy": 0.9083516597747803,
|
|
"step": 120
|
|
},
|
|
{
|
|
"epoch": 0.5627906976744186,
|
|
"grad_norm": 2.06095290184021,
|
|
"learning_rate": 5.390136562446482e-06,
|
|
"loss": 0.2275,
|
|
"mean_token_accuracy": 0.9167689085006714,
|
|
"step": 121
|
|
},
|
|
{
|
|
"epoch": 0.5674418604651162,
|
|
"grad_norm": 2.4146745204925537,
|
|
"learning_rate": 5.316926614470063e-06,
|
|
"loss": 0.3051,
|
|
"mean_token_accuracy": 0.8924481272697449,
|
|
"step": 122
|
|
},
|
|
{
|
|
"epoch": 0.5720930232558139,
|
|
"grad_norm": 2.705653190612793,
|
|
"learning_rate": 5.2437651730843165e-06,
|
|
"loss": 0.3418,
|
|
"mean_token_accuracy": 0.8777977228164673,
|
|
"step": 123
|
|
},
|
|
{
|
|
"epoch": 0.5767441860465117,
|
|
"grad_norm": 2.5144317150115967,
|
|
"learning_rate": 5.170671622931527e-06,
|
|
"loss": 0.3105,
|
|
"mean_token_accuracy": 0.8910664916038513,
|
|
"step": 124
|
|
},
|
|
{
|
|
"epoch": 0.5813953488372093,
|
|
"grad_norm": 2.083824634552002,
|
|
"learning_rate": 5.097665330665714e-06,
|
|
"loss": 0.2399,
|
|
"mean_token_accuracy": 0.9148097634315491,
|
|
"step": 125
|
|
},
|
|
{
|
|
"epoch": 0.586046511627907,
|
|
"grad_norm": 2.6404523849487305,
|
|
"learning_rate": 5.024765639821305e-06,
|
|
"loss": 0.3211,
|
|
"mean_token_accuracy": 0.8938697576522827,
|
|
"step": 126
|
|
},
|
|
{
|
|
"epoch": 0.5906976744186047,
|
|
"grad_norm": 2.0765397548675537,
|
|
"learning_rate": 4.951991865687923e-06,
|
|
"loss": 0.2523,
|
|
"mean_token_accuracy": 0.9152514934539795,
|
|
"step": 127
|
|
},
|
|
{
|
|
"epoch": 0.5953488372093023,
|
|
"grad_norm": 2.412893772125244,
|
|
"learning_rate": 4.879363290192667e-06,
|
|
"loss": 0.2936,
|
|
"mean_token_accuracy": 0.9007983803749084,
|
|
"step": 128
|
|
},
|
|
{
|
|
"epoch": 0.6,
|
|
"grad_norm": 2.7524521350860596,
|
|
"learning_rate": 4.806899156791231e-06,
|
|
"loss": 0.3579,
|
|
"mean_token_accuracy": 0.8833458423614502,
|
|
"step": 129
|
|
},
|
|
{
|
|
"epoch": 0.6046511627906976,
|
|
"grad_norm": 2.421870470046997,
|
|
"learning_rate": 4.734618665369202e-06,
|
|
"loss": 0.2835,
|
|
"mean_token_accuracy": 0.8964264988899231,
|
|
"step": 130
|
|
},
|
|
{
|
|
"epoch": 0.6093023255813953,
|
|
"grad_norm": 2.338674306869507,
|
|
"learning_rate": 4.662540967154934e-06,
|
|
"loss": 0.2663,
|
|
"mean_token_accuracy": 0.9057571887969971,
|
|
"step": 131
|
|
},
|
|
{
|
|
"epoch": 0.6139534883720931,
|
|
"grad_norm": 2.144547700881958,
|
|
"learning_rate": 4.5906851596452765e-06,
|
|
"loss": 0.2723,
|
|
"mean_token_accuracy": 0.9027295112609863,
|
|
"step": 132
|
|
},
|
|
{
|
|
"epoch": 0.6186046511627907,
|
|
"grad_norm": 2.2873592376708984,
|
|
"learning_rate": 4.519070281545571e-06,
|
|
"loss": 0.305,
|
|
"mean_token_accuracy": 0.895431637763977,
|
|
"step": 133
|
|
},
|
|
{
|
|
"epoch": 0.6232558139534884,
|
|
"grad_norm": 2.3061864376068115,
|
|
"learning_rate": 4.447715307725197e-06,
|
|
"loss": 0.3126,
|
|
"mean_token_accuracy": 0.8983712196350098,
|
|
"step": 134
|
|
},
|
|
{
|
|
"epoch": 0.627906976744186,
|
|
"grad_norm": 2.276719093322754,
|
|
"learning_rate": 4.376639144190061e-06,
|
|
"loss": 0.2996,
|
|
"mean_token_accuracy": 0.8973622918128967,
|
|
"step": 135
|
|
},
|
|
{
|
|
"epoch": 0.6325581395348837,
|
|
"grad_norm": 2.5244691371917725,
|
|
"learning_rate": 4.305860623073304e-06,
|
|
"loss": 0.2969,
|
|
"mean_token_accuracy": 0.8997734189033508,
|
|
"step": 136
|
|
},
|
|
{
|
|
"epoch": 0.6372093023255814,
|
|
"grad_norm": 2.136345386505127,
|
|
"learning_rate": 4.2353984976456e-06,
|
|
"loss": 0.2782,
|
|
"mean_token_accuracy": 0.9054514765739441,
|
|
"step": 137
|
|
},
|
|
{
|
|
"epoch": 0.641860465116279,
|
|
"grad_norm": 2.2209346294403076,
|
|
"learning_rate": 4.1652714373463435e-06,
|
|
"loss": 0.2881,
|
|
"mean_token_accuracy": 0.8994898200035095,
|
|
"step": 138
|
|
},
|
|
{
|
|
"epoch": 0.6465116279069767,
|
|
"grad_norm": 2.2153260707855225,
|
|
"learning_rate": 4.095498022837051e-06,
|
|
"loss": 0.2713,
|
|
"mean_token_accuracy": 0.9085735082626343,
|
|
"step": 139
|
|
},
|
|
{
|
|
"epoch": 0.6511627906976745,
|
|
"grad_norm": 2.323110818862915,
|
|
"learning_rate": 4.026096741078281e-06,
|
|
"loss": 0.3132,
|
|
"mean_token_accuracy": 0.8948206901550293,
|
|
"step": 140
|
|
},
|
|
{
|
|
"epoch": 0.6558139534883721,
|
|
"grad_norm": 2.3423662185668945,
|
|
"learning_rate": 3.957085980431382e-06,
|
|
"loss": 0.3104,
|
|
"mean_token_accuracy": 0.8916748762130737,
|
|
"step": 141
|
|
},
|
|
{
|
|
"epoch": 0.6604651162790698,
|
|
"grad_norm": 2.168367624282837,
|
|
"learning_rate": 3.888484025786364e-06,
|
|
"loss": 0.2807,
|
|
"mean_token_accuracy": 0.9040858745574951,
|
|
"step": 142
|
|
},
|
|
{
|
|
"epoch": 0.6651162790697674,
|
|
"grad_norm": 2.1336045265197754,
|
|
"learning_rate": 3.820309053717195e-06,
|
|
"loss": 0.2629,
|
|
"mean_token_accuracy": 0.9082328081130981,
|
|
"step": 143
|
|
},
|
|
{
|
|
"epoch": 0.6697674418604651,
|
|
"grad_norm": 2.4182586669921875,
|
|
"learning_rate": 3.75257912766578e-06,
|
|
"loss": 0.3521,
|
|
"mean_token_accuracy": 0.8834901452064514,
|
|
"step": 144
|
|
},
|
|
{
|
|
"epoch": 0.6744186046511628,
|
|
"grad_norm": 2.415916681289673,
|
|
"learning_rate": 3.6853121931559334e-06,
|
|
"loss": 0.3177,
|
|
"mean_token_accuracy": 0.8882511258125305,
|
|
"step": 145
|
|
},
|
|
{
|
|
"epoch": 0.6790697674418604,
|
|
"grad_norm": 2.250131368637085,
|
|
"learning_rate": 3.618526073038574e-06,
|
|
"loss": 0.3391,
|
|
"mean_token_accuracy": 0.8773726224899292,
|
|
"step": 146
|
|
},
|
|
{
|
|
"epoch": 0.6837209302325581,
|
|
"grad_norm": 2.3493268489837646,
|
|
"learning_rate": 3.552238462769446e-06,
|
|
"loss": 0.2808,
|
|
"mean_token_accuracy": 0.8990680575370789,
|
|
"step": 147
|
|
},
|
|
{
|
|
"epoch": 0.6883720930232559,
|
|
"grad_norm": 2.5086095333099365,
|
|
"learning_rate": 3.4864669257205745e-06,
|
|
"loss": 0.2714,
|
|
"mean_token_accuracy": 0.9056432843208313,
|
|
"step": 148
|
|
},
|
|
{
|
|
"epoch": 0.6930232558139535,
|
|
"grad_norm": 2.324185848236084,
|
|
"learning_rate": 3.4212288885267246e-06,
|
|
"loss": 0.3179,
|
|
"mean_token_accuracy": 0.8921183347702026,
|
|
"step": 149
|
|
},
|
|
{
|
|
"epoch": 0.6976744186046512,
|
|
"grad_norm": 2.469515323638916,
|
|
"learning_rate": 3.3565416364681016e-06,
|
|
"loss": 0.2866,
|
|
"mean_token_accuracy": 0.8951537013053894,
|
|
"step": 150
|
|
},
|
|
{
|
|
"epoch": 0.7023255813953488,
|
|
"grad_norm": 2.097574234008789,
|
|
"learning_rate": 3.2924223088904816e-06,
|
|
"loss": 0.2751,
|
|
"mean_token_accuracy": 0.903618574142456,
|
|
"step": 151
|
|
},
|
|
{
|
|
"epoch": 0.7069767441860465,
|
|
"grad_norm": 2.559396743774414,
|
|
"learning_rate": 3.228887894664029e-06,
|
|
"loss": 0.3377,
|
|
"mean_token_accuracy": 0.8870507478713989,
|
|
"step": 152
|
|
},
|
|
{
|
|
"epoch": 0.7116279069767442,
|
|
"grad_norm": 2.4009549617767334,
|
|
"learning_rate": 3.1659552276819693e-06,
|
|
"loss": 0.3098,
|
|
"mean_token_accuracy": 0.8919164538383484,
|
|
"step": 153
|
|
},
|
|
{
|
|
"epoch": 0.7162790697674418,
|
|
"grad_norm": 2.2265982627868652,
|
|
"learning_rate": 3.1036409824003324e-06,
|
|
"loss": 0.2853,
|
|
"mean_token_accuracy": 0.9013279676437378,
|
|
"step": 154
|
|
},
|
|
{
|
|
"epoch": 0.7209302325581395,
|
|
"grad_norm": 2.1032068729400635,
|
|
"learning_rate": 3.0419616694199327e-06,
|
|
"loss": 0.2629,
|
|
"mean_token_accuracy": 0.9076212644577026,
|
|
"step": 155
|
|
},
|
|
{
|
|
"epoch": 0.7255813953488373,
|
|
"grad_norm": 2.1827595233917236,
|
|
"learning_rate": 2.98093363111178e-06,
|
|
"loss": 0.2401,
|
|
"mean_token_accuracy": 0.915657639503479,
|
|
"step": 156
|
|
},
|
|
{
|
|
"epoch": 0.7302325581395349,
|
|
"grad_norm": 2.498218297958374,
|
|
"learning_rate": 2.92057303728704e-06,
|
|
"loss": 0.2965,
|
|
"mean_token_accuracy": 0.8975441455841064,
|
|
"step": 157
|
|
},
|
|
{
|
|
"epoch": 0.7348837209302326,
|
|
"grad_norm": 2.230644941329956,
|
|
"learning_rate": 2.860895880912735e-06,
|
|
"loss": 0.2833,
|
|
"mean_token_accuracy": 0.9061956405639648,
|
|
"step": 158
|
|
},
|
|
{
|
|
"epoch": 0.7395348837209302,
|
|
"grad_norm": 2.2900495529174805,
|
|
"learning_rate": 2.801917973874294e-06,
|
|
"loss": 0.3014,
|
|
"mean_token_accuracy": 0.8936989903450012,
|
|
"step": 159
|
|
},
|
|
{
|
|
"epoch": 0.7441860465116279,
|
|
"grad_norm": 2.3364789485931396,
|
|
"learning_rate": 2.7436549427860766e-06,
|
|
"loss": 0.3207,
|
|
"mean_token_accuracy": 0.8891250491142273,
|
|
"step": 160
|
|
},
|
|
{
|
|
"epoch": 0.7488372093023256,
|
|
"grad_norm": 2.0765552520751953,
|
|
"learning_rate": 2.6861222248509926e-06,
|
|
"loss": 0.2473,
|
|
"mean_token_accuracy": 0.9146341681480408,
|
|
"step": 161
|
|
},
|
|
{
|
|
"epoch": 0.7534883720930232,
|
|
"grad_norm": 2.1344990730285645,
|
|
"learning_rate": 2.6293350637703123e-06,
|
|
"loss": 0.2804,
|
|
"mean_token_accuracy": 0.9104732871055603,
|
|
"step": 162
|
|
},
|
|
{
|
|
"epoch": 0.7581395348837209,
|
|
"grad_norm": 2.4207024574279785,
|
|
"learning_rate": 2.5733085057047325e-06,
|
|
"loss": 0.2856,
|
|
"mean_token_accuracy": 0.9033816456794739,
|
|
"step": 163
|
|
},
|
|
{
|
|
"epoch": 0.7627906976744186,
|
|
"grad_norm": 2.183412790298462,
|
|
"learning_rate": 2.518057395287792e-06,
|
|
"loss": 0.2971,
|
|
"mean_token_accuracy": 0.8962974548339844,
|
|
"step": 164
|
|
},
|
|
{
|
|
"epoch": 0.7674418604651163,
|
|
"grad_norm": 2.357823371887207,
|
|
"learning_rate": 2.463596371692681e-06,
|
|
"loss": 0.2952,
|
|
"mean_token_accuracy": 0.8953551650047302,
|
|
"step": 165
|
|
},
|
|
{
|
|
"epoch": 0.772093023255814,
|
|
"grad_norm": 2.134320020675659,
|
|
"learning_rate": 2.409939864753487e-06,
|
|
"loss": 0.2963,
|
|
"mean_token_accuracy": 0.8957446813583374,
|
|
"step": 166
|
|
},
|
|
{
|
|
"epoch": 0.7767441860465116,
|
|
"grad_norm": 2.194366693496704,
|
|
"learning_rate": 2.3571020911419067e-06,
|
|
"loss": 0.2818,
|
|
"mean_token_accuracy": 0.9092063307762146,
|
|
"step": 167
|
|
},
|
|
{
|
|
"epoch": 0.7813953488372093,
|
|
"grad_norm": 2.407562494277954,
|
|
"learning_rate": 2.3050970506004463e-06,
|
|
"loss": 0.3105,
|
|
"mean_token_accuracy": 0.8934277892112732,
|
|
"step": 168
|
|
},
|
|
{
|
|
"epoch": 0.786046511627907,
|
|
"grad_norm": 2.3943395614624023,
|
|
"learning_rate": 2.2539385222330797e-06,
|
|
"loss": 0.2516,
|
|
"mean_token_accuracy": 0.9115543961524963,
|
|
"step": 169
|
|
},
|
|
{
|
|
"epoch": 0.7906976744186046,
|
|
"grad_norm": 2.2870237827301025,
|
|
"learning_rate": 2.203640060854387e-06,
|
|
"loss": 0.2662,
|
|
"mean_token_accuracy": 0.9091401696205139,
|
|
"step": 170
|
|
},
|
|
{
|
|
"epoch": 0.7953488372093023,
|
|
"grad_norm": 2.2317891120910645,
|
|
"learning_rate": 2.1542149933981014e-06,
|
|
"loss": 0.3036,
|
|
"mean_token_accuracy": 0.8948342800140381,
|
|
"step": 171
|
|
},
|
|
{
|
|
"epoch": 0.8,
|
|
"grad_norm": 2.2114291191101074,
|
|
"learning_rate": 2.10567641538605e-06,
|
|
"loss": 0.2777,
|
|
"mean_token_accuracy": 0.9104036092758179,
|
|
"step": 172
|
|
},
|
|
{
|
|
"epoch": 0.8046511627906977,
|
|
"grad_norm": 2.1394577026367188,
|
|
"learning_rate": 2.058037187458398e-06,
|
|
"loss": 0.2885,
|
|
"mean_token_accuracy": 0.9054663777351379,
|
|
"step": 173
|
|
},
|
|
{
|
|
"epoch": 0.8093023255813954,
|
|
"grad_norm": 2.3979647159576416,
|
|
"learning_rate": 2.011309931966136e-06,
|
|
"loss": 0.2822,
|
|
"mean_token_accuracy": 0.9056000113487244,
|
|
"step": 174
|
|
},
|
|
{
|
|
"epoch": 0.813953488372093,
|
|
"grad_norm": 2.212559461593628,
|
|
"learning_rate": 1.965507029626695e-06,
|
|
"loss": 0.2933,
|
|
"mean_token_accuracy": 0.8865057826042175,
|
|
"step": 175
|
|
},
|
|
{
|
|
"epoch": 0.8186046511627907,
|
|
"grad_norm": 2.2449798583984375,
|
|
"learning_rate": 1.920640616243589e-06,
|
|
"loss": 0.2605,
|
|
"mean_token_accuracy": 0.9103293418884277,
|
|
"step": 176
|
|
},
|
|
{
|
|
"epoch": 0.8232558139534883,
|
|
"grad_norm": 2.4525561332702637,
|
|
"learning_rate": 1.8767225794909484e-06,
|
|
"loss": 0.318,
|
|
"mean_token_accuracy": 0.8924676179885864,
|
|
"step": 177
|
|
},
|
|
{
|
|
"epoch": 0.827906976744186,
|
|
"grad_norm": 2.1186583042144775,
|
|
"learning_rate": 1.8337645557637929e-06,
|
|
"loss": 0.2723,
|
|
"mean_token_accuracy": 0.9039152264595032,
|
|
"step": 178
|
|
},
|
|
{
|
|
"epoch": 0.8325581395348837,
|
|
"grad_norm": 2.124532461166382,
|
|
"learning_rate": 1.7917779270948887e-06,
|
|
"loss": 0.2916,
|
|
"mean_token_accuracy": 0.8994188904762268,
|
|
"step": 179
|
|
},
|
|
{
|
|
"epoch": 0.8372093023255814,
|
|
"grad_norm": 2.302385091781616,
|
|
"learning_rate": 1.7507738181390027e-06,
|
|
"loss": 0.2811,
|
|
"mean_token_accuracy": 0.9010390639305115,
|
|
"step": 180
|
|
},
|
|
{
|
|
"epoch": 0.8418604651162791,
|
|
"grad_norm": 2.309089422225952,
|
|
"learning_rate": 1.7107630932253383e-06,
|
|
"loss": 0.2813,
|
|
"mean_token_accuracy": 0.902031660079956,
|
|
"step": 181
|
|
},
|
|
{
|
|
"epoch": 0.8465116279069768,
|
|
"grad_norm": 2.244872808456421,
|
|
"learning_rate": 1.6717563534789594e-06,
|
|
"loss": 0.2882,
|
|
"mean_token_accuracy": 0.9040014147758484,
|
|
"step": 182
|
|
},
|
|
{
|
|
"epoch": 0.8511627906976744,
|
|
"grad_norm": 2.278420925140381,
|
|
"learning_rate": 1.6337639340119476e-06,
|
|
"loss": 0.2939,
|
|
"mean_token_accuracy": 0.8955157995223999,
|
|
"step": 183
|
|
},
|
|
{
|
|
"epoch": 0.8558139534883721,
|
|
"grad_norm": 2.336935043334961,
|
|
"learning_rate": 1.596795901185037e-06,
|
|
"loss": 0.3015,
|
|
"mean_token_accuracy": 0.8964377641677856,
|
|
"step": 184
|
|
},
|
|
{
|
|
"epoch": 0.8604651162790697,
|
|
"grad_norm": 2.0233819484710693,
|
|
"learning_rate": 1.5608620499404628e-06,
|
|
"loss": 0.274,
|
|
"mean_token_accuracy": 0.9038158059120178,
|
|
"step": 185
|
|
},
|
|
{
|
|
"epoch": 0.8651162790697674,
|
|
"grad_norm": 2.278059959411621,
|
|
"learning_rate": 1.5259719012067249e-06,
|
|
"loss": 0.2998,
|
|
"mean_token_accuracy": 0.9001963138580322,
|
|
"step": 186
|
|
},
|
|
{
|
|
"epoch": 0.8697674418604651,
|
|
"grad_norm": 2.2441978454589844,
|
|
"learning_rate": 1.4921346993759453e-06,
|
|
"loss": 0.2667,
|
|
"mean_token_accuracy": 0.9050288796424866,
|
|
"step": 187
|
|
},
|
|
{
|
|
"epoch": 0.8744186046511628,
|
|
"grad_norm": 2.1383931636810303,
|
|
"learning_rate": 1.459359409854505e-06,
|
|
"loss": 0.2815,
|
|
"mean_token_accuracy": 0.908123791217804,
|
|
"step": 188
|
|
},
|
|
{
|
|
"epoch": 0.8790697674418605,
|
|
"grad_norm": 2.204313278198242,
|
|
"learning_rate": 1.4276547166875946e-06,
|
|
"loss": 0.2606,
|
|
"mean_token_accuracy": 0.9115758538246155,
|
|
"step": 189
|
|
},
|
|
{
|
|
"epoch": 0.8837209302325582,
|
|
"grad_norm": 2.4099416732788086,
|
|
"learning_rate": 1.397029020258313e-06,
|
|
"loss": 0.2828,
|
|
"mean_token_accuracy": 0.8990963697433472,
|
|
"step": 190
|
|
},
|
|
{
|
|
"epoch": 0.8883720930232558,
|
|
"grad_norm": 2.332491159439087,
|
|
"learning_rate": 1.367490435061928e-06,
|
|
"loss": 0.3299,
|
|
"mean_token_accuracy": 0.8829391598701477,
|
|
"step": 191
|
|
},
|
|
{
|
|
"epoch": 0.8930232558139535,
|
|
"grad_norm": 2.145528554916382,
|
|
"learning_rate": 1.3390467875558855e-06,
|
|
"loss": 0.2444,
|
|
"mean_token_accuracy": 0.9146341681480408,
|
|
"step": 192
|
|
},
|
|
{
|
|
"epoch": 0.8976744186046511,
|
|
"grad_norm": 2.165822982788086,
|
|
"learning_rate": 1.3117056140861317e-06,
|
|
"loss": 0.2863,
|
|
"mean_token_accuracy": 0.9044266939163208,
|
|
"step": 193
|
|
},
|
|
{
|
|
"epoch": 0.9023255813953488,
|
|
"grad_norm": 2.1664891242980957,
|
|
"learning_rate": 1.285474158890304e-06,
|
|
"loss": 0.2596,
|
|
"mean_token_accuracy": 0.9129547476768494,
|
|
"step": 194
|
|
},
|
|
{
|
|
"epoch": 0.9069767441860465,
|
|
"grad_norm": 2.0730209350585938,
|
|
"learning_rate": 1.2603593721783219e-06,
|
|
"loss": 0.2618,
|
|
"mean_token_accuracy": 0.9128862023353577,
|
|
"step": 195
|
|
},
|
|
{
|
|
"epoch": 0.9116279069767442,
|
|
"grad_norm": 2.156203508377075,
|
|
"learning_rate": 1.2363679082908766e-06,
|
|
"loss": 0.2953,
|
|
"mean_token_accuracy": 0.8979016542434692,
|
|
"step": 196
|
|
},
|
|
{
|
|
"epoch": 0.9162790697674419,
|
|
"grad_norm": 2.236882209777832,
|
|
"learning_rate": 1.2135061239363161e-06,
|
|
"loss": 0.2538,
|
|
"mean_token_accuracy": 0.9129107594490051,
|
|
"step": 197
|
|
},
|
|
{
|
|
"epoch": 0.9209302325581395,
|
|
"grad_norm": 2.2554514408111572,
|
|
"learning_rate": 1.1917800765063954e-06,
|
|
"loss": 0.2668,
|
|
"mean_token_accuracy": 0.9089072346687317,
|
|
"step": 198
|
|
},
|
|
{
|
|
"epoch": 0.9255813953488372,
|
|
"grad_norm": 2.4423582553863525,
|
|
"learning_rate": 1.1711955224713209e-06,
|
|
"loss": 0.2917,
|
|
"mean_token_accuracy": 0.898510217666626,
|
|
"step": 199
|
|
},
|
|
{
|
|
"epoch": 0.9302325581395349,
|
|
"grad_norm": 2.198429822921753,
|
|
"learning_rate": 1.1517579158545386e-06,
|
|
"loss": 0.2671,
|
|
"mean_token_accuracy": 0.9085792303085327,
|
|
"step": 200
|
|
},
|
|
{
|
|
"epoch": 0.9348837209302325,
|
|
"grad_norm": 2.3745462894439697,
|
|
"learning_rate": 1.1334724067876463e-06,
|
|
"loss": 0.2967,
|
|
"mean_token_accuracy": 0.9004044532775879,
|
|
"step": 201
|
|
},
|
|
{
|
|
"epoch": 0.9395348837209302,
|
|
"grad_norm": 2.1694419384002686,
|
|
"learning_rate": 1.1163438401458358e-06,
|
|
"loss": 0.263,
|
|
"mean_token_accuracy": 0.906028687953949,
|
|
"step": 202
|
|
},
|
|
{
|
|
"epoch": 0.9441860465116279,
|
|
"grad_norm": 2.3387255668640137,
|
|
"learning_rate": 1.1003767542642021e-06,
|
|
"loss": 0.2795,
|
|
"mean_token_accuracy": 0.9063853025436401,
|
|
"step": 203
|
|
},
|
|
{
|
|
"epoch": 0.9488372093023256,
|
|
"grad_norm": 2.1712303161621094,
|
|
"learning_rate": 1.0855753797352868e-06,
|
|
"loss": 0.2498,
|
|
"mean_token_accuracy": 0.9100555777549744,
|
|
"step": 204
|
|
},
|
|
{
|
|
"epoch": 0.9534883720930233,
|
|
"grad_norm": 2.0819292068481445,
|
|
"learning_rate": 1.0719436382881466e-06,
|
|
"loss": 0.2587,
|
|
"mean_token_accuracy": 0.9077077507972717,
|
|
"step": 205
|
|
},
|
|
{
|
|
"epoch": 0.958139534883721,
|
|
"grad_norm": 2.2681405544281006,
|
|
"learning_rate": 1.0594851417492665e-06,
|
|
"loss": 0.28,
|
|
"mean_token_accuracy": 0.9018029570579529,
|
|
"step": 206
|
|
},
|
|
{
|
|
"epoch": 0.9627906976744186,
|
|
"grad_norm": 2.4561564922332764,
|
|
"learning_rate": 1.0482031910855804e-06,
|
|
"loss": 0.2997,
|
|
"mean_token_accuracy": 0.8940787315368652,
|
|
"step": 207
|
|
},
|
|
{
|
|
"epoch": 0.9674418604651163,
|
|
"grad_norm": 2.234903335571289,
|
|
"learning_rate": 1.0381007755298547e-06,
|
|
"loss": 0.2715,
|
|
"mean_token_accuracy": 0.9086244106292725,
|
|
"step": 208
|
|
},
|
|
{
|
|
"epoch": 0.9720930232558139,
|
|
"grad_norm": 2.2118217945098877,
|
|
"learning_rate": 1.029180571788672e-06,
|
|
"loss": 0.2549,
|
|
"mean_token_accuracy": 0.9135593175888062,
|
|
"step": 209
|
|
},
|
|
{
|
|
"epoch": 0.9767441860465116,
|
|
"grad_norm": 2.0356454849243164,
|
|
"learning_rate": 1.021444943333218e-06,
|
|
"loss": 0.2619,
|
|
"mean_token_accuracy": 0.9112606048583984,
|
|
"step": 210
|
|
},
|
|
{
|
|
"epoch": 0.9813953488372092,
|
|
"grad_norm": 2.2461442947387695,
|
|
"learning_rate": 1.0148959397730637e-06,
|
|
"loss": 0.2587,
|
|
"mean_token_accuracy": 0.9082584977149963,
|
|
"step": 211
|
|
},
|
|
{
|
|
"epoch": 0.986046511627907,
|
|
"grad_norm": 2.1244120597839355,
|
|
"learning_rate": 1.0095352963131057e-06,
|
|
"loss": 0.2805,
|
|
"mean_token_accuracy": 0.9018282294273376,
|
|
"step": 212
|
|
},
|
|
{
|
|
"epoch": 0.9906976744186047,
|
|
"grad_norm": 2.2568986415863037,
|
|
"learning_rate": 1.0053644332938118e-06,
|
|
"loss": 0.2488,
|
|
"mean_token_accuracy": 0.9099595546722412,
|
|
"step": 213
|
|
},
|
|
{
|
|
"epoch": 0.9953488372093023,
|
|
"grad_norm": 2.1964008808135986,
|
|
"learning_rate": 1.0023844558148912e-06,
|
|
"loss": 0.295,
|
|
"mean_token_accuracy": 0.8972173929214478,
|
|
"step": 214
|
|
},
|
|
{
|
|
"epoch": 1.0,
|
|
"grad_norm": 2.0329501628875732,
|
|
"learning_rate": 1.0005961534424925e-06,
|
|
"loss": 0.2276,
|
|
"mean_token_accuracy": 0.9119472503662109,
|
|
"step": 215
|
|
},
|
|
{
|
|
"epoch": 1.0,
|
|
"step": 215,
|
|
"total_flos": 2.0144468407196058e+17,
|
|
"train_loss": 0.32494123903817906,
|
|
"train_runtime": 1259.521,
|
|
"train_samples_per_second": 5.439,
|
|
"train_steps_per_second": 0.171
|
|
}
|
|
],
|
|
"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
|
|
}
|