{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9798270893371758,
  "eval_steps": 10,
  "global_step": 340,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01440922190201729,
      "grad_norm": 0.888121485710144,
      "learning_rate": 0.0002988472622478386,
      "loss": 2.4115,
      "step": 5
    },
    {
      "epoch": 0.02881844380403458,
      "grad_norm": 0.8970298767089844,
      "learning_rate": 0.00029740634005763684,
      "loss": 2.23,
      "step": 10
    },
    {
      "epoch": 0.02881844380403458,
      "eval_loss": 2.165903091430664,
      "eval_runtime": 1.8407,
      "eval_samples_per_second": 84.207,
      "eval_steps_per_second": 10.865,
      "step": 10
    },
    {
      "epoch": 0.043227665706051875,
      "grad_norm": 1.078068494796753,
      "learning_rate": 0.00029596541786743513,
      "loss": 2.0155,
      "step": 15
    },
    {
      "epoch": 0.05763688760806916,
      "grad_norm": 1.2069385051727295,
      "learning_rate": 0.0002945244956772334,
      "loss": 1.9346,
      "step": 20
    },
    {
      "epoch": 0.05763688760806916,
      "eval_loss": 1.7943660020828247,
      "eval_runtime": 1.7809,
      "eval_samples_per_second": 87.035,
      "eval_steps_per_second": 11.23,
      "step": 20
    },
    {
      "epoch": 0.07204610951008646,
      "grad_norm": 1.3408102989196777,
      "learning_rate": 0.0002930835734870317,
      "loss": 1.7434,
      "step": 25
    },
    {
      "epoch": 0.08645533141210375,
      "grad_norm": 1.4849472045898438,
      "learning_rate": 0.00029164265129682994,
      "loss": 1.5409,
      "step": 30
    },
    {
      "epoch": 0.08645533141210375,
      "eval_loss": 1.5225533246994019,
      "eval_runtime": 1.7657,
      "eval_samples_per_second": 87.784,
      "eval_steps_per_second": 11.327,
      "step": 30
    },
    {
      "epoch": 0.10086455331412104,
      "grad_norm": 2.2489142417907715,
      "learning_rate": 0.0002902017291066282,
      "loss": 1.4702,
      "step": 35
    },
    {
      "epoch": 0.11527377521613832,
      "grad_norm": 2.169492244720459,
      "learning_rate": 0.00028876080691642647,
      "loss": 1.398,
      "step": 40
    },
    {
      "epoch": 0.11527377521613832,
      "eval_loss": 1.3084412813186646,
      "eval_runtime": 1.7715,
      "eval_samples_per_second": 87.495,
      "eval_steps_per_second": 11.29,
      "step": 40
    },
    {
      "epoch": 0.12968299711815562,
      "grad_norm": 2.665741205215454,
      "learning_rate": 0.00028731988472622475,
      "loss": 1.2839,
      "step": 45
    },
    {
      "epoch": 0.1440922190201729,
      "grad_norm": 2.9897077083587646,
      "learning_rate": 0.00028587896253602304,
      "loss": 1.0893,
      "step": 50
    },
    {
      "epoch": 0.1440922190201729,
      "eval_loss": 1.0865856409072876,
      "eval_runtime": 1.7767,
      "eval_samples_per_second": 87.241,
      "eval_steps_per_second": 11.257,
      "step": 50
    },
    {
      "epoch": 0.1585014409221902,
      "grad_norm": 2.3380894660949707,
      "learning_rate": 0.0002844380403458213,
      "loss": 1.0454,
      "step": 55
    },
    {
      "epoch": 0.1729106628242075,
      "grad_norm": 2.571589469909668,
      "learning_rate": 0.00028299711815561957,
      "loss": 0.8605,
      "step": 60
    },
    {
      "epoch": 0.1729106628242075,
      "eval_loss": 0.8586989641189575,
      "eval_runtime": 1.7853,
      "eval_samples_per_second": 86.822,
      "eval_steps_per_second": 11.203,
      "step": 60
    },
    {
      "epoch": 0.1873198847262248,
      "grad_norm": 3.0408358573913574,
      "learning_rate": 0.00028155619596541786,
      "loss": 0.8101,
      "step": 65
    },
    {
      "epoch": 0.2017291066282421,
      "grad_norm": 2.4773292541503906,
      "learning_rate": 0.0002801152737752161,
      "loss": 0.7769,
      "step": 70
    },
    {
      "epoch": 0.2017291066282421,
      "eval_loss": 0.6477732062339783,
      "eval_runtime": 1.781,
      "eval_samples_per_second": 87.029,
      "eval_steps_per_second": 11.23,
      "step": 70
    },
    {
      "epoch": 0.21613832853025935,
      "grad_norm": 2.5505621433258057,
      "learning_rate": 0.0002786743515850144,
      "loss": 0.6286,
      "step": 75
    },
    {
      "epoch": 0.23054755043227665,
      "grad_norm": 2.5216686725616455,
      "learning_rate": 0.00027723342939481267,
      "loss": 0.5306,
      "step": 80
    },
    {
      "epoch": 0.23054755043227665,
      "eval_loss": 0.4904349446296692,
      "eval_runtime": 1.7712,
      "eval_samples_per_second": 87.51,
      "eval_steps_per_second": 11.292,
      "step": 80
    },
    {
      "epoch": 0.24495677233429394,
      "grad_norm": 3.0844411849975586,
      "learning_rate": 0.00027579250720461096,
      "loss": 0.5331,
      "step": 85
    },
    {
      "epoch": 0.25936599423631124,
      "grad_norm": 1.8952299356460571,
      "learning_rate": 0.0002743515850144092,
      "loss": 0.4093,
      "step": 90
    },
    {
      "epoch": 0.25936599423631124,
      "eval_loss": 0.40096166729927063,
      "eval_runtime": 1.773,
      "eval_samples_per_second": 87.422,
      "eval_steps_per_second": 11.28,
      "step": 90
    },
    {
      "epoch": 0.2737752161383285,
      "grad_norm": 3.3445639610290527,
      "learning_rate": 0.0002729106628242075,
      "loss": 0.3654,
      "step": 95
    },
    {
      "epoch": 0.2881844380403458,
      "grad_norm": 1.9506555795669556,
      "learning_rate": 0.0002714697406340057,
      "loss": 0.3458,
      "step": 100
    },
    {
      "epoch": 0.2881844380403458,
      "eval_loss": 0.32525885105133057,
      "eval_runtime": 1.7918,
      "eval_samples_per_second": 86.503,
      "eval_steps_per_second": 11.162,
      "step": 100
    },
    {
      "epoch": 0.3025936599423631,
      "grad_norm": 1.9951375722885132,
      "learning_rate": 0.000270028818443804,
      "loss": 0.2672,
      "step": 105
    },
    {
      "epoch": 0.3170028818443804,
      "grad_norm": 2.8618788719177246,
      "learning_rate": 0.0002685878962536023,
      "loss": 0.3316,
      "step": 110
    },
    {
      "epoch": 0.3170028818443804,
      "eval_loss": 0.29092785716056824,
      "eval_runtime": 1.7704,
      "eval_samples_per_second": 87.549,
      "eval_steps_per_second": 11.297,
      "step": 110
    },
    {
      "epoch": 0.3314121037463977,
      "grad_norm": 2.436544179916382,
      "learning_rate": 0.00026714697406340053,
      "loss": 0.3176,
      "step": 115
    },
    {
      "epoch": 0.345821325648415,
      "grad_norm": 1.1800215244293213,
      "learning_rate": 0.0002657060518731988,
      "loss": 0.2378,
      "step": 120
    },
    {
      "epoch": 0.345821325648415,
      "eval_loss": 0.25983747839927673,
      "eval_runtime": 1.7719,
      "eval_samples_per_second": 87.476,
      "eval_steps_per_second": 11.287,
      "step": 120
    },
    {
      "epoch": 0.36023054755043227,
      "grad_norm": 1.0937371253967285,
      "learning_rate": 0.0002642651296829971,
      "loss": 0.2617,
      "step": 125
    },
    {
      "epoch": 0.3746397694524496,
      "grad_norm": 1.5132169723510742,
      "learning_rate": 0.0002628242074927954,
      "loss": 0.2669,
      "step": 130
    },
    {
      "epoch": 0.3746397694524496,
      "eval_loss": 0.2434806078672409,
      "eval_runtime": 1.7851,
      "eval_samples_per_second": 86.831,
      "eval_steps_per_second": 11.204,
      "step": 130
    },
    {
      "epoch": 0.38904899135446686,
      "grad_norm": 1.4011964797973633,
      "learning_rate": 0.00026138328530259363,
      "loss": 0.2684,
      "step": 135
    },
    {
      "epoch": 0.4034582132564842,
      "grad_norm": 1.3246668577194214,
      "learning_rate": 0.0002599423631123919,
      "loss": 0.2377,
      "step": 140
    },
    {
      "epoch": 0.4034582132564842,
      "eval_loss": 0.2348952293395996,
      "eval_runtime": 1.7807,
      "eval_samples_per_second": 87.042,
      "eval_steps_per_second": 11.231,
      "step": 140
    },
    {
      "epoch": 0.41786743515850144,
      "grad_norm": 2.753978967666626,
      "learning_rate": 0.0002585014409221902,
      "loss": 0.2716,
      "step": 145
    },
    {
      "epoch": 0.4322766570605187,
      "grad_norm": 0.8502065539360046,
      "learning_rate": 0.00025706051873198844,
      "loss": 0.2282,
      "step": 150
    },
    {
      "epoch": 0.4322766570605187,
      "eval_loss": 0.23007912933826447,
      "eval_runtime": 1.766,
      "eval_samples_per_second": 87.771,
      "eval_steps_per_second": 11.325,
      "step": 150
    },
    {
      "epoch": 0.44668587896253603,
      "grad_norm": 1.0703223943710327,
      "learning_rate": 0.00025561959654178673,
      "loss": 0.2374,
      "step": 155
    },
    {
      "epoch": 0.4610951008645533,
      "grad_norm": 0.7980679869651794,
      "learning_rate": 0.00025417867435158497,
      "loss": 0.2151,
      "step": 160
    },
    {
      "epoch": 0.4610951008645533,
      "eval_loss": 0.23066848516464233,
      "eval_runtime": 1.7821,
      "eval_samples_per_second": 86.978,
      "eval_steps_per_second": 11.223,
      "step": 160
    },
    {
      "epoch": 0.4755043227665706,
      "grad_norm": 1.1535905599594116,
      "learning_rate": 0.00025273775216138326,
      "loss": 0.2086,
      "step": 165
    },
    {
      "epoch": 0.4899135446685879,
      "grad_norm": 0.9484102725982666,
      "learning_rate": 0.00025129682997118155,
      "loss": 0.2127,
      "step": 170
    },
    {
      "epoch": 0.4899135446685879,
      "eval_loss": 0.2209121137857437,
      "eval_runtime": 1.7727,
      "eval_samples_per_second": 87.439,
      "eval_steps_per_second": 11.283,
      "step": 170
    },
    {
      "epoch": 0.5043227665706052,
      "grad_norm": 0.9559063911437988,
      "learning_rate": 0.0002498559077809798,
      "loss": 0.2134,
      "step": 175
    },
    {
      "epoch": 0.5187319884726225,
      "grad_norm": 0.66960209608078,
      "learning_rate": 0.00024841498559077807,
      "loss": 0.2555,
      "step": 180
    },
    {
      "epoch": 0.5187319884726225,
      "eval_loss": 0.21855449676513672,
      "eval_runtime": 1.7887,
      "eval_samples_per_second": 86.654,
      "eval_steps_per_second": 11.181,
      "step": 180
    },
    {
      "epoch": 0.5331412103746398,
      "grad_norm": 0.6968249082565308,
      "learning_rate": 0.00024697406340057636,
      "loss": 0.2199,
      "step": 185
    },
    {
      "epoch": 0.547550432276657,
      "grad_norm": 0.6100601553916931,
      "learning_rate": 0.00024553314121037465,
      "loss": 0.2348,
      "step": 190
    },
    {
      "epoch": 0.547550432276657,
      "eval_loss": 0.21677546203136444,
      "eval_runtime": 1.7784,
      "eval_samples_per_second": 87.156,
      "eval_steps_per_second": 11.246,
      "step": 190
    },
    {
      "epoch": 0.5619596541786743,
      "grad_norm": 0.6942987442016602,
      "learning_rate": 0.00024409221902017288,
      "loss": 0.2165,
      "step": 195
    },
    {
      "epoch": 0.5763688760806917,
      "grad_norm": 0.8180645704269409,
      "learning_rate": 0.00024265129682997117,
      "loss": 0.2141,
      "step": 200
    },
    {
      "epoch": 0.5763688760806917,
      "eval_loss": 0.21652507781982422,
      "eval_runtime": 1.7751,
      "eval_samples_per_second": 87.319,
      "eval_steps_per_second": 11.267,
      "step": 200
    },
    {
      "epoch": 0.590778097982709,
      "grad_norm": 0.6298684477806091,
      "learning_rate": 0.00024121037463976943,
      "loss": 0.2261,
      "step": 205
    },
    {
      "epoch": 0.6051873198847262,
      "grad_norm": 2.6468467712402344,
      "learning_rate": 0.0002397694524495677,
      "loss": 0.245,
      "step": 210
    },
    {
      "epoch": 0.6051873198847262,
      "eval_loss": 0.2188318818807602,
      "eval_runtime": 1.7737,
      "eval_samples_per_second": 87.389,
      "eval_steps_per_second": 11.276,
      "step": 210
    },
    {
      "epoch": 0.6195965417867435,
      "grad_norm": 0.6621644496917725,
      "learning_rate": 0.00023832853025936598,
      "loss": 0.1866,
      "step": 215
    },
    {
      "epoch": 0.6340057636887608,
      "grad_norm": 0.5494632124900818,
      "learning_rate": 0.00023688760806916425,
      "loss": 0.2128,
      "step": 220
    },
    {
      "epoch": 0.6340057636887608,
      "eval_loss": 0.21284270286560059,
      "eval_runtime": 1.787,
      "eval_samples_per_second": 86.739,
      "eval_steps_per_second": 11.192,
      "step": 220
    },
    {
      "epoch": 0.6484149855907781,
      "grad_norm": 0.5896772146224976,
      "learning_rate": 0.00023544668587896253,
      "loss": 0.2266,
      "step": 225
    },
    {
      "epoch": 0.6628242074927954,
      "grad_norm": 0.45764079689979553,
      "learning_rate": 0.0002340057636887608,
      "loss": 0.2393,
      "step": 230
    },
    {
      "epoch": 0.6628242074927954,
      "eval_loss": 0.21253199875354767,
      "eval_runtime": 1.7769,
      "eval_samples_per_second": 87.232,
      "eval_steps_per_second": 11.256,
      "step": 230
    },
    {
      "epoch": 0.6772334293948127,
      "grad_norm": 0.699612557888031,
      "learning_rate": 0.00023256484149855909,
      "loss": 0.2517,
      "step": 235
    },
    {
      "epoch": 0.69164265129683,
      "grad_norm": 0.42438164353370667,
      "learning_rate": 0.00023112391930835732,
      "loss": 0.2016,
      "step": 240
    },
    {
      "epoch": 0.69164265129683,
      "eval_loss": 0.21218827366828918,
      "eval_runtime": 1.7865,
      "eval_samples_per_second": 86.762,
      "eval_steps_per_second": 11.195,
      "step": 240
    },
    {
      "epoch": 0.7060518731988472,
      "grad_norm": 0.5191032886505127,
      "learning_rate": 0.00022968299711815558,
      "loss": 0.2218,
      "step": 245
    },
    {
      "epoch": 0.7204610951008645,
      "grad_norm": 0.5536476969718933,
      "learning_rate": 0.00022824207492795387,
      "loss": 0.2145,
      "step": 250
    },
    {
      "epoch": 0.7204610951008645,
      "eval_loss": 0.2105206400156021,
      "eval_runtime": 1.8096,
      "eval_samples_per_second": 85.654,
      "eval_steps_per_second": 11.052,
      "step": 250
    },
    {
      "epoch": 0.7348703170028819,
      "grad_norm": 0.5729750394821167,
      "learning_rate": 0.00022680115273775213,
      "loss": 0.2356,
      "step": 255
    },
    {
      "epoch": 0.7492795389048992,
      "grad_norm": 0.442891389131546,
      "learning_rate": 0.00022536023054755042,
      "loss": 0.203,
      "step": 260
    },
    {
      "epoch": 0.7492795389048992,
      "eval_loss": 0.20932228863239288,
      "eval_runtime": 1.77,
      "eval_samples_per_second": 87.57,
      "eval_steps_per_second": 11.299,
      "step": 260
    },
    {
      "epoch": 0.7636887608069164,
      "grad_norm": 0.556273341178894,
      "learning_rate": 0.00022391930835734868,
      "loss": 0.2092,
      "step": 265
    },
    {
      "epoch": 0.7780979827089337,
      "grad_norm": 0.461923211812973,
      "learning_rate": 0.00022247838616714695,
      "loss": 0.2116,
      "step": 270
    },
    {
      "epoch": 0.7780979827089337,
      "eval_loss": 0.2080857753753662,
      "eval_runtime": 1.7792,
      "eval_samples_per_second": 87.12,
      "eval_steps_per_second": 11.241,
      "step": 270
    },
    {
      "epoch": 0.792507204610951,
      "grad_norm": 0.5841118693351746,
      "learning_rate": 0.00022103746397694523,
      "loss": 0.2244,
      "step": 275
    },
    {
      "epoch": 0.8069164265129684,
      "grad_norm": 0.5412226319313049,
      "learning_rate": 0.0002195965417867435,
      "loss": 0.1943,
      "step": 280
    },
    {
      "epoch": 0.8069164265129684,
      "eval_loss": 0.208485409617424,
      "eval_runtime": 1.7916,
      "eval_samples_per_second": 86.515,
      "eval_steps_per_second": 11.163,
      "step": 280
    },
    {
      "epoch": 0.8213256484149856,
      "grad_norm": 0.5236246585845947,
      "learning_rate": 0.00021815561959654179,
      "loss": 0.2243,
      "step": 285
    },
    {
      "epoch": 0.8357348703170029,
      "grad_norm": 0.48271429538726807,
      "learning_rate": 0.00021671469740634002,
      "loss": 0.2123,
      "step": 290
    },
    {
      "epoch": 0.8357348703170029,
      "eval_loss": 0.20807716250419617,
      "eval_runtime": 1.7753,
      "eval_samples_per_second": 87.308,
      "eval_steps_per_second": 11.266,
      "step": 290
    },
    {
      "epoch": 0.8501440922190202,
      "grad_norm": 0.4914911985397339,
      "learning_rate": 0.0002152737752161383,
      "loss": 0.2252,
      "step": 295
    },
    {
      "epoch": 0.8645533141210374,
      "grad_norm": 0.46419402956962585,
      "learning_rate": 0.00021383285302593657,
      "loss": 0.1999,
      "step": 300
    },
    {
      "epoch": 0.8645533141210374,
      "eval_loss": 0.20786339044570923,
      "eval_runtime": 1.7845,
      "eval_samples_per_second": 86.858,
      "eval_steps_per_second": 11.208,
      "step": 300
    },
    {
      "epoch": 0.8789625360230547,
      "grad_norm": 0.540306568145752,
      "learning_rate": 0.00021239193083573483,
      "loss": 0.1856,
      "step": 305
    },
    {
      "epoch": 0.8933717579250721,
      "grad_norm": 0.46772050857543945,
      "learning_rate": 0.00021095100864553312,
      "loss": 0.2185,
      "step": 310
    },
    {
      "epoch": 0.8933717579250721,
      "eval_loss": 0.20842696726322174,
      "eval_runtime": 1.7693,
      "eval_samples_per_second": 87.604,
      "eval_steps_per_second": 11.304,
      "step": 310
    },
    {
      "epoch": 0.9077809798270894,
      "grad_norm": 0.5399373173713684,
      "learning_rate": 0.00020951008645533138,
      "loss": 0.2108,
      "step": 315
    },
    {
      "epoch": 0.9221902017291066,
      "grad_norm": 0.5167156457901001,
      "learning_rate": 0.00020806916426512967,
      "loss": 0.2364,
      "step": 320
    },
    {
      "epoch": 0.9221902017291066,
      "eval_loss": 0.2061736136674881,
      "eval_runtime": 1.7797,
      "eval_samples_per_second": 87.095,
      "eval_steps_per_second": 11.238,
      "step": 320
    },
    {
      "epoch": 0.9365994236311239,
      "grad_norm": 0.5894590616226196,
      "learning_rate": 0.00020662824207492793,
      "loss": 0.2185,
      "step": 325
    },
    {
      "epoch": 0.9510086455331412,
      "grad_norm": 0.4573725163936615,
      "learning_rate": 0.00020518731988472622,
      "loss": 0.2109,
      "step": 330
    },
    {
      "epoch": 0.9510086455331412,
      "eval_loss": 0.20550554990768433,
      "eval_runtime": 1.787,
      "eval_samples_per_second": 86.736,
      "eval_steps_per_second": 11.192,
      "step": 330
    },
    {
      "epoch": 0.9654178674351584,
      "grad_norm": 0.4973134696483612,
      "learning_rate": 0.00020374639769452449,
      "loss": 0.2354,
      "step": 335
    },
    {
      "epoch": 0.9798270893371758,
      "grad_norm": 0.5064740180969238,
      "learning_rate": 0.00020230547550432275,
      "loss": 0.2263,
      "step": 340
    },
    {
      "epoch": 0.9798270893371758,
      "eval_loss": 0.20669177174568176,
      "eval_runtime": 1.7751,
      "eval_samples_per_second": 87.32,
      "eval_steps_per_second": 11.267,
      "step": 340
    }
  ],
  "logging_steps": 5,
  "max_steps": 1041,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4326815988449280.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}