{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9813084112149533,
  "eval_steps": 26,
  "global_step": 424,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02336448598130841,
      "grad_norm": 16.625,
      "learning_rate": 1.0477479818042947e-05,
      "loss": 1.0571476936340332,
      "step": 5
    },
    {
      "epoch": 0.04672897196261682,
      "grad_norm": 5.84375,
      "learning_rate": 2.357432959059663e-05,
      "loss": 0.5670325279235839,
      "step": 10
    },
    {
      "epoch": 0.07009345794392523,
      "grad_norm": 4.34375,
      "learning_rate": 2.6192615944772643e-05,
      "loss": 0.534730339050293,
      "step": 15
    },
    {
      "epoch": 0.09345794392523364,
      "grad_norm": 5.03125,
      "learning_rate": 2.6188214228158065e-05,
      "loss": 0.528727674484253,
      "step": 20
    },
    {
      "epoch": 0.11682242990654206,
      "grad_norm": 4.84375,
      "learning_rate": 2.618042818650518e-05,
      "loss": 0.5030421733856201,
      "step": 25
    },
    {
      "epoch": 0.14018691588785046,
      "grad_norm": 5.1875,
      "learning_rate": 2.6169260503942314e-05,
      "loss": 0.5157054424285888,
      "step": 30
    },
    {
      "epoch": 0.16355140186915887,
      "grad_norm": 4.09375,
      "learning_rate": 2.6154715030370898e-05,
      "loss": 0.4850313186645508,
      "step": 35
    },
    {
      "epoch": 0.18691588785046728,
      "grad_norm": 4.5,
      "learning_rate": 2.6136796780138316e-05,
      "loss": 0.4838898181915283,
      "step": 40
    },
    {
      "epoch": 0.2102803738317757,
      "grad_norm": 4.25,
      "learning_rate": 2.6115511930309236e-05,
      "loss": 0.5209589004516602,
      "step": 45
    },
    {
      "epoch": 0.2336448598130841,
      "grad_norm": 4.375,
      "learning_rate": 2.609086781853618e-05,
      "loss": 0.48980093002319336,
      "step": 50
    },
    {
      "epoch": 0.24299065420560748,
      "eval_loss": 0.4843875467777252,
      "eval_runtime": 55.6497,
      "eval_samples_per_second": 7.116,
      "eval_steps_per_second": 7.116,
      "step": 52
    },
    {
      "epoch": 0.24766355140186916,
      "eval_loss": 0.4850113093852997,
      "eval_runtime": 63.4929,
      "eval_samples_per_second": 6.237,
      "eval_steps_per_second": 6.237,
      "step": 53
    },
    {
      "epoch": 0.2570093457943925,
      "grad_norm": 4.28125,
      "learning_rate": 2.606287294052997e-05,
      "loss": 0.46539983749389646,
      "step": 55
    },
    {
      "epoch": 0.2803738317757009,
      "grad_norm": 4.4375,
      "learning_rate": 2.6031536947130925e-05,
      "loss": 0.4871384143829346,
      "step": 60
    },
    {
      "epoch": 0.3037383177570093,
      "grad_norm": 3.890625,
      "learning_rate": 2.599687064098191e-05,
      "loss": 0.4708998203277588,
      "step": 65
    },
    {
      "epoch": 0.32710280373831774,
      "grad_norm": 4.0625,
      "learning_rate": 2.5958885972804244e-05,
      "loss": 0.4961503028869629,
      "step": 70
    },
    {
      "epoch": 0.35046728971962615,
      "grad_norm": 4.0625,
      "learning_rate": 2.591759603727788e-05,
      "loss": 0.4586769104003906,
      "step": 75
    },
    {
      "epoch": 0.3644859813084112,
      "eval_loss": 0.47235846519470215,
      "eval_runtime": 60.6823,
      "eval_samples_per_second": 6.526,
      "eval_steps_per_second": 6.526,
      "step": 78
    },
    {
      "epoch": 0.37383177570093457,
      "grad_norm": 4.3125,
      "learning_rate": 2.587301506852717e-05,
      "loss": 0.45466132164001466,
      "step": 80
    },
    {
      "epoch": 0.397196261682243,
      "grad_norm": 3.90625,
      "learning_rate": 2.582515843521389e-05,
      "loss": 0.45511479377746583,
      "step": 85
    },
    {
      "epoch": 0.4205607476635514,
      "grad_norm": 4.28125,
      "learning_rate": 2.577404263523907e-05,
      "loss": 0.44046764373779296,
      "step": 90
    },
    {
      "epoch": 0.4439252336448598,
      "grad_norm": 3.65625,
      "learning_rate": 2.57196852900556e-05,
      "loss": 0.4845860958099365,
      "step": 95
    },
    {
      "epoch": 0.4672897196261682,
      "grad_norm": 3.34375,
      "learning_rate": 2.566210513859348e-05,
      "loss": 0.46938457489013674,
      "step": 100
    },
    {
      "epoch": 0.48598130841121495,
      "eval_loss": 0.4602718651294708,
      "eval_runtime": 57.4013,
      "eval_samples_per_second": 6.899,
      "eval_steps_per_second": 6.899,
      "step": 104
    },
    {
      "epoch": 0.49065420560747663,
      "grad_norm": 4.25,
      "learning_rate": 2.5601322030799827e-05,
      "loss": 0.4705543041229248,
      "step": 105
    },
    {
      "epoch": 0.4953271028037383,
      "eval_loss": 0.45894455909729004,
      "eval_runtime": 64.3876,
      "eval_samples_per_second": 6.15,
      "eval_steps_per_second": 6.15,
      "step": 106
    },
    {
      "epoch": 0.514018691588785,
      "grad_norm": 3.8125,
      "learning_rate": 2.5537356920795893e-05,
      "loss": 0.45661373138427735,
      "step": 110
    },
    {
      "epoch": 0.5373831775700935,
      "grad_norm": 3.96875,
      "learning_rate": 2.5470231859653427e-05,
      "loss": 0.4881335735321045,
      "step": 115
    },
    {
      "epoch": 0.5607476635514018,
      "grad_norm": 3.5625,
      "learning_rate": 2.539996998779286e-05,
      "loss": 0.4613192558288574,
      "step": 120
    },
    {
      "epoch": 0.5841121495327103,
      "grad_norm": 3.96875,
      "learning_rate": 2.532659552700601e-05,
      "loss": 0.482680082321167,
      "step": 125
    },
    {
      "epoch": 0.6074766355140186,
      "grad_norm": 3.84375,
      "learning_rate": 2.525013377210592e-05,
      "loss": 0.4225574493408203,
      "step": 130
    },
    {
      "epoch": 0.6074766355140186,
      "eval_loss": 0.45120301842689514,
      "eval_runtime": 64.3659,
      "eval_samples_per_second": 6.152,
      "eval_steps_per_second": 6.152,
      "step": 130
    },
    {
      "epoch": 0.6308411214953271,
      "grad_norm": 4.09375,
      "learning_rate": 2.517061108220685e-05,
      "loss": 0.43398637771606446,
      "step": 135
    },
    {
      "epoch": 0.6542056074766355,
      "grad_norm": 3.65625,
      "learning_rate": 2.5088054871637335e-05,
      "loss": 0.4436498165130615,
      "step": 140
    },
    {
      "epoch": 0.677570093457944,
      "grad_norm": 3.5,
      "learning_rate": 2.500249360048947e-05,
      "loss": 0.44930062294006345,
      "step": 145
    },
    {
      "epoch": 0.7009345794392523,
      "grad_norm": 3.796875,
      "learning_rate": 2.4913956764807738e-05,
      "loss": 0.45390973091125486,
      "step": 150
    },
    {
      "epoch": 0.7242990654205608,
      "grad_norm": 4.15625,
      "learning_rate": 2.4822474886420593e-05,
      "loss": 0.4333805561065674,
      "step": 155
    },
    {
      "epoch": 0.7289719626168224,
      "eval_loss": 0.44806569814682007,
      "eval_runtime": 58.3512,
      "eval_samples_per_second": 6.786,
      "eval_steps_per_second": 6.786,
      "step": 156
    },
    {
      "epoch": 0.7429906542056075,
      "eval_loss": 0.44679105281829834,
      "eval_runtime": 53.2561,
      "eval_samples_per_second": 7.436,
      "eval_steps_per_second": 7.436,
      "step": 159
    },
    {
      "epoch": 0.7476635514018691,
      "grad_norm": 3.28125,
      "learning_rate": 2.4728079502418555e-05,
      "loss": 0.4298356056213379,
      "step": 160
    },
    {
      "epoch": 0.7710280373831776,
      "grad_norm": 3.5,
      "learning_rate": 2.46308031542822e-05,
      "loss": 0.45917558670043945,
      "step": 165
    },
    {
      "epoch": 0.794392523364486,
      "grad_norm": 3.40625,
      "learning_rate": 2.4530679376663966e-05,
      "loss": 0.4135223388671875,
      "step": 170
    },
    {
      "epoch": 0.8177570093457944,
      "grad_norm": 3.5625,
      "learning_rate": 2.4427742685827546e-05,
      "loss": 0.4308349609375,
      "step": 175
    },
    {
      "epoch": 0.8411214953271028,
      "grad_norm": 3.5625,
      "learning_rate": 2.432202856774887e-05,
      "loss": 0.4360647678375244,
      "step": 180
    },
    {
      "epoch": 0.8504672897196262,
      "eval_loss": 0.44622713327407837,
      "eval_runtime": 54.0702,
      "eval_samples_per_second": 7.324,
      "eval_steps_per_second": 7.324,
      "step": 182
    },
    {
      "epoch": 0.8644859813084113,
      "grad_norm": 3.40625,
      "learning_rate": 2.421357346588286e-05,
      "loss": 0.44407052993774415,
      "step": 185
    },
    {
      "epoch": 0.8878504672897196,
      "grad_norm": 3.515625,
      "learning_rate": 2.410241476860002e-05,
      "loss": 0.42552666664123534,
      "step": 190
    },
    {
      "epoch": 0.9112149532710281,
      "grad_norm": 3.90625,
      "learning_rate": 2.3988590796297346e-05,
      "loss": 0.47596588134765627,
      "step": 195
    },
    {
      "epoch": 0.9345794392523364,
      "grad_norm": 3.84375,
      "learning_rate": 2.3872140788187892e-05,
      "loss": 0.4219630241394043,
      "step": 200
    },
    {
      "epoch": 0.9579439252336449,
      "grad_norm": 3.453125,
      "learning_rate": 2.37531048887736e-05,
      "loss": 0.44322934150695803,
      "step": 205
    },
    {
      "epoch": 0.9719626168224299,
      "eval_loss": 0.442614883184433,
      "eval_runtime": 62.8656,
      "eval_samples_per_second": 6.299,
      "eval_steps_per_second": 6.299,
      "step": 208
    },
    {
      "epoch": 0.9813084112149533,
      "grad_norm": 3.375,
      "learning_rate": 2.3631524134006092e-05,
      "loss": 0.4280299186706543,
      "step": 210
    },
    {
      "epoch": 0.9906542056074766,
      "eval_loss": 0.43994140625,
      "eval_runtime": 57.058,
      "eval_samples_per_second": 6.94,
      "eval_steps_per_second": 6.94,
      "step": 212
    },
    {
      "epoch": 1.0046728971962617,
      "grad_norm": 2.78125,
      "learning_rate": 2.350744043714004e-05,
      "loss": 0.412522029876709,
      "step": 215
    },
    {
      "epoch": 1.02803738317757,
      "grad_norm": 3.265625,
      "learning_rate": 2.3380896574284195e-05,
      "loss": 0.3000824213027954,
      "step": 220
    },
    {
      "epoch": 1.0514018691588785,
      "grad_norm": 3.625,
      "learning_rate": 2.3251936169654928e-05,
      "loss": 0.29865827560424807,
      "step": 225
    },
    {
      "epoch": 1.074766355140187,
      "grad_norm": 3.234375,
      "learning_rate": 2.3120603680537385e-05,
      "loss": 0.32170352935791013,
      "step": 230
    },
    {
      "epoch": 1.0934579439252337,
      "eval_loss": 0.4519166946411133,
      "eval_runtime": 60.0729,
      "eval_samples_per_second": 6.592,
      "eval_steps_per_second": 6.592,
      "step": 234
    },
    {
      "epoch": 1.0981308411214954,
      "grad_norm": 3.5,
      "learning_rate": 2.2986944381959467e-05,
      "loss": 0.32140297889709474,
      "step": 235
    },
    {
      "epoch": 1.1214953271028036,
      "grad_norm": 3.546875,
      "learning_rate": 2.2851004351083894e-05,
      "loss": 0.3222970485687256,
      "step": 240
    },
    {
      "epoch": 1.144859813084112,
      "grad_norm": 3.234375,
      "learning_rate": 2.271283045132377e-05,
      "loss": 0.31093592643737794,
      "step": 245
    },
    {
      "epoch": 1.1682242990654206,
      "grad_norm": 3.65625,
      "learning_rate": 2.2572470316187064e-05,
      "loss": 0.3200420141220093,
      "step": 250
    },
    {
      "epoch": 1.191588785046729,
      "grad_norm": 3.34375,
      "learning_rate": 2.2429972332855617e-05,
      "loss": 0.3097134351730347,
      "step": 255
    },
    {
      "epoch": 1.2149532710280373,
      "grad_norm": 3.4375,
      "learning_rate": 2.2285385625504386e-05,
      "loss": 0.2976833820343018,
      "step": 260
    },
    {
      "epoch": 1.2149532710280373,
      "eval_loss": 0.45616593956947327,
      "eval_runtime": 56.6532,
      "eval_samples_per_second": 6.99,
      "eval_steps_per_second": 6.99,
      "step": 260
    },
    {
      "epoch": 1.2383177570093458,
      "grad_norm": 3.453125,
      "learning_rate": 2.213876003836653e-05,
      "loss": 0.302638578414917,
      "step": 265
    },
    {
      "epoch": 1.2383177570093458,
      "eval_loss": 0.4536404013633728,
      "eval_runtime": 63.9718,
      "eval_samples_per_second": 6.19,
      "eval_steps_per_second": 6.19,
      "step": 265
    },
    {
      "epoch": 1.2616822429906542,
      "grad_norm": 3.1875,
      "learning_rate": 2.1990146118550325e-05,
      "loss": 0.3010557651519775,
      "step": 270
    },
    {
      "epoch": 1.2850467289719627,
      "grad_norm": 3.515625,
      "learning_rate": 2.183959509861378e-05,
      "loss": 0.29446218013763426,
      "step": 275
    },
    {
      "epoch": 1.308411214953271,
      "grad_norm": 3.546875,
      "learning_rate": 2.16871588789029e-05,
      "loss": 0.3248613357543945,
      "step": 280
    },
    {
      "epoch": 1.3317757009345794,
      "grad_norm": 3.96875,
      "learning_rate": 2.1532890009659805e-05,
      "loss": 0.29060137271881104,
      "step": 285
    },
    {
      "epoch": 1.3364485981308412,
      "eval_loss": 0.4469919800758362,
      "eval_runtime": 40.3935,
      "eval_samples_per_second": 9.804,
      "eval_steps_per_second": 9.804,
      "step": 286
    },
    {
      "epoch": 1.355140186915888,
      "grad_norm": 3.453125,
      "learning_rate": 2.137684167290675e-05,
      "loss": 0.30477614402770997,
      "step": 290
    },
    {
      "epoch": 1.3785046728971961,
      "grad_norm": 3.46875,
      "learning_rate": 2.1219067664112372e-05,
      "loss": 0.31689789295196535,
      "step": 295
    },
    {
      "epoch": 1.4018691588785046,
      "grad_norm": 3.65625,
      "learning_rate": 2.1059622373646478e-05,
      "loss": 0.28146681785583494,
      "step": 300
    },
    {
      "epoch": 1.425233644859813,
      "grad_norm": 3.5,
      "learning_rate": 2.0898560768029707e-05,
      "loss": 0.3086798667907715,
      "step": 305
    },
    {
      "epoch": 1.4485981308411215,
      "grad_norm": 3.53125,
      "learning_rate": 2.0735938370984625e-05,
      "loss": 0.30053484439849854,
      "step": 310
    },
    {
      "epoch": 1.4579439252336448,
      "eval_loss": 0.4480014443397522,
      "eval_runtime": 47.0762,
      "eval_samples_per_second": 8.412,
      "eval_steps_per_second": 8.412,
      "step": 312
    },
    {
      "epoch": 1.47196261682243,
      "grad_norm": 3.59375,
      "learning_rate": 2.0571811244294655e-05,
      "loss": 0.30966882705688475,
      "step": 315
    },
    {
      "epoch": 1.485981308411215,
      "eval_loss": 0.4486345648765564,
      "eval_runtime": 50.9547,
      "eval_samples_per_second": 7.772,
      "eval_steps_per_second": 7.772,
      "step": 318
    },
    {
      "epoch": 1.4953271028037383,
      "grad_norm": 3.296875,
      "learning_rate": 2.040623596847762e-05,
      "loss": 0.30327141284942627,
      "step": 320
    },
    {
      "epoch": 1.5186915887850467,
      "grad_norm": 3.15625,
      "learning_rate": 2.0239269623280375e-05,
      "loss": 0.2928351879119873,
      "step": 325
    },
    {
      "epoch": 1.542056074766355,
      "grad_norm": 3.640625,
      "learning_rate": 2.0070969768001355e-05,
      "loss": 0.30496037006378174,
      "step": 330
    },
    {
      "epoch": 1.5654205607476634,
      "grad_norm": 3.375,
      "learning_rate": 1.9901394421647857e-05,
      "loss": 0.30050930976867674,
      "step": 335
    },
    {
      "epoch": 1.5794392523364484,
      "eval_loss": 0.4455008804798126,
      "eval_runtime": 65.8944,
      "eval_samples_per_second": 6.01,
      "eval_steps_per_second": 6.01,
      "step": 338
    },
    {
      "epoch": 1.588785046728972,
      "grad_norm": 3.640625,
      "learning_rate": 1.9730602042934775e-05,
      "loss": 0.30058119297027586,
      "step": 340
    },
    {
      "epoch": 1.6121495327102804,
      "grad_norm": 3.390625,
      "learning_rate": 1.955865151013178e-05,
      "loss": 0.3002913951873779,
      "step": 345
    },
    {
      "epoch": 1.6355140186915889,
      "grad_norm": 3.65625,
      "learning_rate": 1.9385602100765877e-05,
      "loss": 0.2978963375091553,
      "step": 350
    },
    {
      "epoch": 1.6588785046728973,
      "grad_norm": 3.421875,
      "learning_rate": 1.921151347118631e-05,
      "loss": 0.30049972534179686,
      "step": 355
    },
    {
      "epoch": 1.6822429906542056,
      "grad_norm": 3.515625,
      "learning_rate": 1.9036445635998852e-05,
      "loss": 0.3050508499145508,
      "step": 360
    },
    {
      "epoch": 1.7009345794392523,
      "eval_loss": 0.4410502016544342,
      "eval_runtime": 64.5037,
      "eval_samples_per_second": 6.139,
      "eval_steps_per_second": 6.139,
      "step": 364
    },
    {
      "epoch": 1.705607476635514,
      "grad_norm": 3.640625,
      "learning_rate": 1.8860458947376653e-05,
      "loss": 0.29273965358734133,
      "step": 365
    },
    {
      "epoch": 1.7289719626168223,
      "grad_norm": 3.296875,
      "learning_rate": 1.868361407425465e-05,
      "loss": 0.2988182544708252,
      "step": 370
    },
    {
      "epoch": 1.7336448598130842,
      "eval_loss": 0.441893070936203,
      "eval_runtime": 57.6479,
      "eval_samples_per_second": 6.869,
      "eval_steps_per_second": 6.869,
      "step": 371
    },
    {
      "epoch": 1.7523364485981308,
      "grad_norm": 3.546875,
      "learning_rate": 1.8505971981414815e-05,
      "loss": 0.2948916912078857,
      "step": 375
    },
    {
      "epoch": 1.7757009345794392,
      "grad_norm": 4.03125,
      "learning_rate": 1.8327593908469397e-05,
      "loss": 0.3248611927032471,
      "step": 380
    },
    {
      "epoch": 1.7990654205607477,
      "grad_norm": 3.375,
      "learning_rate": 1.814854134874943e-05,
      "loss": 0.3255984544754028,
      "step": 385
    },
    {
      "epoch": 1.8224299065420562,
      "grad_norm": 3.859375,
      "learning_rate": 1.796887602810579e-05,
      "loss": 0.29858739376068116,
      "step": 390
    },
    {
      "epoch": 1.8224299065420562,
      "eval_loss": 0.43862152099609375,
      "eval_runtime": 61.1595,
      "eval_samples_per_second": 6.475,
      "eval_steps_per_second": 6.475,
      "step": 390
    },
    {
      "epoch": 1.8457943925233646,
      "grad_norm": 3.609375,
      "learning_rate": 1.7788659883630035e-05,
      "loss": 0.31806967258453367,
      "step": 395
    },
    {
      "epoch": 1.8691588785046729,
      "grad_norm": 3.703125,
      "learning_rate": 1.760795504230251e-05,
      "loss": 0.3062360525131226,
      "step": 400
    },
    {
      "epoch": 1.8925233644859814,
      "grad_norm": 3.90625,
      "learning_rate": 1.742682379957492e-05,
      "loss": 0.30818881988525393,
      "step": 405
    },
    {
      "epoch": 1.9158878504672896,
      "grad_norm": 3.71875,
      "learning_rate": 1.7245328597894847e-05,
      "loss": 0.2948075532913208,
      "step": 410
    },
    {
      "epoch": 1.939252336448598,
      "grad_norm": 3.265625,
      "learning_rate": 1.7063532005179608e-05,
      "loss": 0.2957978487014771,
      "step": 415
    },
    {
      "epoch": 1.9439252336448598,
      "eval_loss": 0.43459007143974304,
      "eval_runtime": 66.1061,
      "eval_samples_per_second": 5.99,
      "eval_steps_per_second": 5.99,
      "step": 416
    },
    {
      "epoch": 1.9626168224299065,
      "grad_norm": 3.84375,
      "learning_rate": 1.688149669324683e-05,
      "loss": 0.3337227821350098,
      "step": 420
    },
    {
      "epoch": 1.9813084112149533,
      "eval_loss": 0.4314902722835541,
      "eval_runtime": 66.515,
      "eval_samples_per_second": 5.954,
      "eval_steps_per_second": 5.954,
      "step": 424
    }
  ],
  "logging_steps": 5,
  "max_steps": 856,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 26,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.9865684657086464e+16,
  "train_batch_size": 35,
  "trial_name": null,
  "trial_params": null
}