{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9993935718617344,
  "eval_steps": 100,
  "global_step": 412,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01212856276531231,
      "grad_norm": 14.8125,
      "learning_rate": 2.380952380952381e-06,
      "loss": 0.8047,
      "mean_token_accuracy": 0.8034533846529813,
      "step": 5
    },
    {
      "epoch": 0.02425712553062462,
      "grad_norm": 6.96875,
      "learning_rate": 4.761904761904762e-06,
      "loss": 0.7637,
      "mean_token_accuracy": 0.8090282869012707,
      "step": 10
    },
    {
      "epoch": 0.036385688295936934,
      "grad_norm": 3.984375,
      "learning_rate": 7.1428571428571436e-06,
      "loss": 0.721,
      "mean_token_accuracy": 0.8129444648093843,
      "step": 15
    },
    {
      "epoch": 0.04851425106124924,
      "grad_norm": 2.90625,
      "learning_rate": 9.523809523809525e-06,
      "loss": 0.6684,
      "mean_token_accuracy": 0.8205217497556208,
      "step": 20
    },
    {
      "epoch": 0.06064281382656155,
      "grad_norm": 2.0625,
      "learning_rate": 1.1904761904761905e-05,
      "loss": 0.6118,
      "mean_token_accuracy": 0.8317586143695014,
      "step": 25
    },
    {
      "epoch": 0.07277137659187387,
      "grad_norm": 1.296875,
      "learning_rate": 1.4285714285714287e-05,
      "loss": 0.6032,
      "mean_token_accuracy": 0.8310453323558162,
      "step": 30
    },
    {
      "epoch": 0.08489993935718618,
      "grad_norm": 1.0703125,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.5922,
      "mean_token_accuracy": 0.8324016373411535,
      "step": 35
    },
    {
      "epoch": 0.09702850212249849,
      "grad_norm": 0.99609375,
      "learning_rate": 1.904761904761905e-05,
      "loss": 0.5777,
      "mean_token_accuracy": 0.8366324535679374,
      "step": 40
    },
    {
      "epoch": 0.1091570648878108,
      "grad_norm": 0.93359375,
      "learning_rate": 1.9996755966513702e-05,
      "loss": 0.584,
      "mean_token_accuracy": 0.8340573069403714,
      "step": 45
    },
    {
      "epoch": 0.1212856276531231,
      "grad_norm": 0.94921875,
      "learning_rate": 1.9976938939060172e-05,
      "loss": 0.5828,
      "mean_token_accuracy": 0.8344712243401758,
      "step": 50
    },
    {
      "epoch": 0.1334141904184354,
      "grad_norm": 0.91796875,
      "learning_rate": 1.99391427932908e-05,
      "loss": 0.5625,
      "mean_token_accuracy": 0.8387799364613879,
      "step": 55
    },
    {
      "epoch": 0.14554275318374774,
      "grad_norm": 1.5625,
      "learning_rate": 1.9883435640414922e-05,
      "loss": 0.565,
      "mean_token_accuracy": 0.8390288978494626,
      "step": 60
    },
    {
      "epoch": 0.15767131594906003,
      "grad_norm": 0.91015625,
      "learning_rate": 1.980991786848731e-05,
      "loss": 0.5705,
      "mean_token_accuracy": 0.8378207478005866,
      "step": 65
    },
    {
      "epoch": 0.16979987871437235,
      "grad_norm": 0.953125,
      "learning_rate": 1.971872196150208e-05,
      "loss": 0.5674,
      "mean_token_accuracy": 0.8382407746823068,
      "step": 70
    },
    {
      "epoch": 0.18192844147968465,
      "grad_norm": 0.8671875,
      "learning_rate": 1.961001226064762e-05,
      "loss": 0.5744,
      "mean_token_accuracy": 0.8358473851417398,
      "step": 75
    },
    {
      "epoch": 0.19405700424499697,
      "grad_norm": 0.8828125,
      "learning_rate": 1.9483984668152618e-05,
      "loss": 0.5809,
      "mean_token_accuracy": 0.8347812805474095,
      "step": 80
    },
    {
      "epoch": 0.20618556701030927,
      "grad_norm": 1.0078125,
      "learning_rate": 1.9340866294257044e-05,
      "loss": 0.5687,
      "mean_token_accuracy": 0.8378635141739978,
      "step": 85
    },
    {
      "epoch": 0.2183141297756216,
      "grad_norm": 6.25,
      "learning_rate": 1.918091504794411e-05,
      "loss": 0.565,
      "mean_token_accuracy": 0.8387218963831866,
      "step": 90
    },
    {
      "epoch": 0.23044269254093389,
      "grad_norm": 0.9296875,
      "learning_rate": 1.9004419172170887e-05,
      "loss": 0.5746,
      "mean_token_accuracy": 0.8359634652981427,
      "step": 95
    },
    {
      "epoch": 0.2425712553062462,
      "grad_norm": 0.921875,
      "learning_rate": 1.881169672443498e-05,
      "loss": 0.5633,
      "mean_token_accuracy": 0.8383049242424242,
      "step": 100
    },
    {
      "epoch": 0.2425712553062462,
      "eval_loss": 0.5801094770431519,
      "eval_mean_token_accuracy": 0.8435259856630825,
      "eval_runtime": 0.5662,
      "eval_samples_per_second": 128.931,
      "eval_steps_per_second": 5.299,
      "step": 100
    },
    {
      "epoch": 0.2546998180715585,
      "grad_norm": 0.8828125,
      "learning_rate": 1.860309500361345e-05,
      "loss": 0.5686,
      "mean_token_accuracy": 0.8384546065493647,
      "step": 105
    },
    {
      "epoch": 0.2668283808368708,
      "grad_norm": 0.921875,
      "learning_rate": 1.837898992410674e-05,
      "loss": 0.5634,
      "mean_token_accuracy": 0.8386195625610948,
      "step": 110
    },
    {
      "epoch": 0.27895694360218315,
      "grad_norm": 0.9375,
      "learning_rate": 1.8139785338415515e-05,
      "loss": 0.5706,
      "mean_token_accuracy": 0.8389448924731184,
      "step": 115
    },
    {
      "epoch": 0.2910855063674955,
      "grad_norm": 0.890625,
      "learning_rate": 1.788591230937119e-05,
      "loss": 0.547,
      "mean_token_accuracy": 0.8426228005865102,
      "step": 120
    },
    {
      "epoch": 0.30321406913280774,
      "grad_norm": 0.91015625,
      "learning_rate": 1.7617828333331547e-05,
      "loss": 0.5609,
      "mean_token_accuracy": 0.8385706867057674,
      "step": 125
    },
    {
      "epoch": 0.31534263189812006,
      "grad_norm": 0.90625,
      "learning_rate": 1.733601651574137e-05,
      "loss": 0.5632,
      "mean_token_accuracy": 0.8384118401759533,
      "step": 130
    },
    {
      "epoch": 0.3274711946634324,
      "grad_norm": 0.8828125,
      "learning_rate": 1.704098470054379e-05,
      "loss": 0.5547,
      "mean_token_accuracy": 0.8413352272727275,
      "step": 135
    },
    {
      "epoch": 0.3395997574287447,
      "grad_norm": 2.921875,
      "learning_rate": 1.6733264555011196e-05,
      "loss": 0.5655,
      "mean_token_accuracy": 0.8374816715542522,
      "step": 140
    },
    {
      "epoch": 0.35172832019405703,
      "grad_norm": 0.91796875,
      "learning_rate": 1.6413410611644827e-05,
      "loss": 0.5737,
      "mean_token_accuracy": 0.8358046187683283,
      "step": 145
    },
    {
      "epoch": 0.3638568829593693,
      "grad_norm": 0.89453125,
      "learning_rate": 1.6081999268869765e-05,
      "loss": 0.5612,
      "mean_token_accuracy": 0.8389051808406647,
      "step": 150
    },
    {
      "epoch": 0.3759854457246816,
      "grad_norm": 0.859375,
      "learning_rate": 1.5739627752325995e-05,
      "loss": 0.5586,
      "mean_token_accuracy": 0.8389387829912023,
      "step": 155
    },
    {
      "epoch": 0.38811400848999394,
      "grad_norm": 0.8828125,
      "learning_rate": 1.538691303862744e-05,
      "loss": 0.5651,
      "mean_token_accuracy": 0.8381796798631476,
      "step": 160
    },
    {
      "epoch": 0.40024257125530627,
      "grad_norm": 0.8515625,
      "learning_rate": 1.5024490743528393e-05,
      "loss": 0.5635,
      "mean_token_accuracy": 0.8385920698924731,
      "step": 165
    },
    {
      "epoch": 0.41237113402061853,
      "grad_norm": 0.890625,
      "learning_rate": 1.4653013976500977e-05,
      "loss": 0.5644,
      "mean_token_accuracy": 0.8381750977517107,
      "step": 170
    },
    {
      "epoch": 0.42449969678593086,
      "grad_norm": 0.88671875,
      "learning_rate": 1.4273152163787726e-05,
      "loss": 0.5645,
      "mean_token_accuracy": 0.8384500244379277,
      "step": 175
    },
    {
      "epoch": 0.4366282595512432,
      "grad_norm": 0.90234375,
      "learning_rate": 1.3885589842050253e-05,
      "loss": 0.562,
      "mean_token_accuracy": 0.8376863391984362,
      "step": 180
    },
    {
      "epoch": 0.4487568223165555,
      "grad_norm": 0.96875,
      "learning_rate": 1.3491025424787916e-05,
      "loss": 0.5687,
      "mean_token_accuracy": 0.8375504032258064,
      "step": 185
    },
    {
      "epoch": 0.46088538508186777,
      "grad_norm": 0.8984375,
      "learning_rate": 1.3090169943749475e-05,
      "loss": 0.571,
      "mean_token_accuracy": 0.8365408113391984,
      "step": 190
    },
    {
      "epoch": 0.4730139478471801,
      "grad_norm": 0.84375,
      "learning_rate": 1.2683745767605846e-05,
      "loss": 0.5591,
      "mean_token_accuracy": 0.83859665200391,
      "step": 195
    },
    {
      "epoch": 0.4851425106124924,
      "grad_norm": 0.90234375,
      "learning_rate": 1.2272485300192902e-05,
      "loss": 0.5463,
      "mean_token_accuracy": 0.8424104960899316,
      "step": 200
    },
    {
      "epoch": 0.4851425106124924,
      "eval_loss": 0.5737597942352295,
      "eval_mean_token_accuracy": 0.8452422975272437,
      "eval_runtime": 0.5658,
      "eval_samples_per_second": 129.014,
      "eval_steps_per_second": 5.302,
      "step": 200
    },
    {
      "epoch": 0.49727107337780474,
      "grad_norm": 0.90234375,
      "learning_rate": 1.1857129660670281e-05,
      "loss": 0.5524,
      "mean_token_accuracy": 0.8409075635386118,
      "step": 205
    },
    {
      "epoch": 0.509399636143117,
      "grad_norm": 0.8671875,
      "learning_rate": 1.1438427347974554e-05,
      "loss": 0.5547,
      "mean_token_accuracy": 0.8406127810361681,
      "step": 210
    },
    {
      "epoch": 0.5215281989084294,
      "grad_norm": 0.921875,
      "learning_rate": 1.101713289197356e-05,
      "loss": 0.5697,
      "mean_token_accuracy": 0.8376405180840665,
      "step": 215
    },
    {
      "epoch": 0.5336567616737417,
      "grad_norm": 0.87890625,
      "learning_rate": 1.0594005493752568e-05,
      "loss": 0.5496,
      "mean_token_accuracy": 0.8417033235581621,
      "step": 220
    },
    {
      "epoch": 0.5457853244390539,
      "grad_norm": 0.8515625,
      "learning_rate": 1.0169807657482625e-05,
      "loss": 0.5546,
      "mean_token_accuracy": 0.8411183406647116,
      "step": 225
    },
    {
      "epoch": 0.5579138872043663,
      "grad_norm": 0.89453125,
      "learning_rate": 9.745303816336488e-06,
      "loss": 0.5588,
      "mean_token_accuracy": 0.8404768450635386,
      "step": 230
    },
    {
      "epoch": 0.5700424499696786,
      "grad_norm": 0.875,
      "learning_rate": 9.321258954928394e-06,
      "loss": 0.548,
      "mean_token_accuracy": 0.8420546187683285,
      "step": 235
    },
    {
      "epoch": 0.582171012734991,
      "grad_norm": 0.88671875,
      "learning_rate": 8.898437230760058e-06,
      "loss": 0.5577,
      "mean_token_accuracy": 0.8416300097751712,
      "step": 240
    },
    {
      "epoch": 0.5942995755003032,
      "grad_norm": 0.87109375,
      "learning_rate": 8.47760059715722e-06,
      "loss": 0.5555,
      "mean_token_accuracy": 0.840709005376344,
      "step": 245
    },
    {
      "epoch": 0.6064281382656155,
      "grad_norm": 0.87109375,
      "learning_rate": 8.059507430178248e-06,
      "loss": 0.5689,
      "mean_token_accuracy": 0.8373686461388076,
      "step": 250
    },
    {
      "epoch": 0.6185567010309279,
      "grad_norm": 0.875,
      "learning_rate": 7.644911161969225e-06,
      "loss": 0.5592,
      "mean_token_accuracy": 0.8399468475073313,
      "step": 255
    },
    {
      "epoch": 0.6306852637962401,
      "grad_norm": 0.8984375,
      "learning_rate": 7.2345589230282895e-06,
      "loss": 0.5643,
      "mean_token_accuracy": 0.8382331378299119,
      "step": 260
    },
    {
      "epoch": 0.6428138265615525,
      "grad_norm": 0.875,
      "learning_rate": 6.829190195825983e-06,
      "loss": 0.5619,
      "mean_token_accuracy": 0.8390609726295211,
      "step": 265
    },
    {
      "epoch": 0.6549423893268648,
      "grad_norm": 0.86328125,
      "learning_rate": 6.429535482207847e-06,
      "loss": 0.5524,
      "mean_token_accuracy": 0.8410572458455523,
      "step": 270
    },
    {
      "epoch": 0.667070952092177,
      "grad_norm": 0.88671875,
      "learning_rate": 6.036314986980749e-06,
      "loss": 0.5562,
      "mean_token_accuracy": 0.8407471896383187,
      "step": 275
    },
    {
      "epoch": 0.6791995148574894,
      "grad_norm": 0.921875,
      "learning_rate": 5.650237320055107e-06,
      "loss": 0.5642,
      "mean_token_accuracy": 0.8386149804496578,
      "step": 280
    },
    {
      "epoch": 0.6913280776228017,
      "grad_norm": 0.89453125,
      "learning_rate": 5.271998219481953e-06,
      "loss": 0.5696,
      "mean_token_accuracy": 0.8364155669599217,
      "step": 285
    },
    {
      "epoch": 0.7034566403881141,
      "grad_norm": 0.8671875,
      "learning_rate": 4.902279297685945e-06,
      "loss": 0.5426,
      "mean_token_accuracy": 0.844026454056696,
      "step": 290
    },
    {
      "epoch": 0.7155852031534263,
      "grad_norm": 0.8828125,
      "learning_rate": 4.541746813153698e-06,
      "loss": 0.5547,
      "mean_token_accuracy": 0.8409136730205278,
      "step": 295
    },
    {
      "epoch": 0.7277137659187386,
      "grad_norm": 0.8671875,
      "learning_rate": 4.191050469790961e-06,
      "loss": 0.5606,
      "mean_token_accuracy": 0.8398429863147605,
      "step": 300
    },
    {
      "epoch": 0.7277137659187386,
      "eval_loss": 0.5722371339797974,
      "eval_mean_token_accuracy": 0.8450929546359655,
      "eval_runtime": 0.5672,
      "eval_samples_per_second": 128.703,
      "eval_steps_per_second": 5.289,
      "step": 300
    },
    {
      "epoch": 0.739842328684051,
      "grad_norm": 0.85546875,
      "learning_rate": 3.8508222461122565e-06,
      "loss": 0.5574,
      "mean_token_accuracy": 0.840337854349951,
      "step": 305
    },
    {
      "epoch": 0.7519708914493632,
      "grad_norm": 0.87109375,
      "learning_rate": 3.5216752563729004e-06,
      "loss": 0.5512,
      "mean_token_accuracy": 0.8418026026392962,
      "step": 310
    },
    {
      "epoch": 0.7640994542146755,
      "grad_norm": 0.8515625,
      "learning_rate": 3.2042026456956554e-06,
      "loss": 0.5604,
      "mean_token_accuracy": 0.8396490102639296,
      "step": 315
    },
    {
      "epoch": 0.7762280169799879,
      "grad_norm": 0.88671875,
      "learning_rate": 2.8989765211831433e-06,
      "loss": 0.5566,
      "mean_token_accuracy": 0.8406647116324534,
      "step": 320
    },
    {
      "epoch": 0.7883565797453002,
      "grad_norm": 0.90625,
      "learning_rate": 2.6065469209421568e-06,
      "loss": 0.5445,
      "mean_token_accuracy": 0.8426793132942327,
      "step": 325
    },
    {
      "epoch": 0.8004851425106125,
      "grad_norm": 0.88671875,
      "learning_rate": 2.3274408228778357e-06,
      "loss": 0.5578,
      "mean_token_accuracy": 0.8400522360703813,
      "step": 330
    },
    {
      "epoch": 0.8126137052759248,
      "grad_norm": 0.859375,
      "learning_rate": 2.062161195043851e-06,
      "loss": 0.5553,
      "mean_token_accuracy": 0.8400812561094819,
      "step": 335
    },
    {
      "epoch": 0.8247422680412371,
      "grad_norm": 0.8828125,
      "learning_rate": 1.8111860892599832e-06,
      "loss": 0.5528,
      "mean_token_accuracy": 0.841089320625611,
      "step": 340
    },
    {
      "epoch": 0.8368708308065494,
      "grad_norm": 0.8984375,
      "learning_rate": 1.574967779630414e-06,
      "loss": 0.5516,
      "mean_token_accuracy": 0.8412099828934508,
      "step": 345
    },
    {
      "epoch": 0.8489993935718617,
      "grad_norm": 0.84765625,
      "learning_rate": 1.3539319475152457e-06,
      "loss": 0.5465,
      "mean_token_accuracy": 0.8418774437927663,
      "step": 350
    },
    {
      "epoch": 0.8611279563371741,
      "grad_norm": 0.8828125,
      "learning_rate": 1.1484769144239038e-06,
      "loss": 0.5579,
      "mean_token_accuracy": 0.8393939393939392,
      "step": 355
    },
    {
      "epoch": 0.8732565191024864,
      "grad_norm": 0.83984375,
      "learning_rate": 9.589729242128531e-07,
      "loss": 0.5568,
      "mean_token_accuracy": 0.8397498167155426,
      "step": 360
    },
    {
      "epoch": 0.8853850818677986,
      "grad_norm": 0.85546875,
      "learning_rate": 7.857614758811527e-07,
      "loss": 0.5573,
      "mean_token_accuracy": 0.8403760386119259,
      "step": 365
    },
    {
      "epoch": 0.897513644633111,
      "grad_norm": 0.875,
      "learning_rate": 6.291547081661631e-07,
      "loss": 0.5459,
      "mean_token_accuracy": 0.8424960288367547,
      "step": 370
    },
    {
      "epoch": 0.9096422073984233,
      "grad_norm": 0.86328125,
      "learning_rate": 4.894348370484648e-07,
      "loss": 0.5585,
      "mean_token_accuracy": 0.8394794721407625,
      "step": 375
    },
    {
      "epoch": 0.9217707701637355,
      "grad_norm": 0.8828125,
      "learning_rate": 3.668536471795614e-07,
      "loss": 0.5671,
      "mean_token_accuracy": 0.8379566837732162,
      "step": 380
    },
    {
      "epoch": 0.9338993329290479,
      "grad_norm": 0.8984375,
      "learning_rate": 2.6163203814894013e-07,
      "loss": 0.5508,
      "mean_token_accuracy": 0.8413230083088953,
      "step": 385
    },
    {
      "epoch": 0.9460278956943602,
      "grad_norm": 0.8828125,
      "learning_rate": 1.7395962640806675e-07,
      "loss": 0.5474,
      "mean_token_accuracy": 0.8426228005865102,
      "step": 390
    },
    {
      "epoch": 0.9581564584596726,
      "grad_norm": 0.89453125,
      "learning_rate": 1.039944035687368e-07,
      "loss": 0.558,
      "mean_token_accuracy": 0.8407914833822092,
      "step": 395
    },
    {
      "epoch": 0.9702850212249848,
      "grad_norm": 0.90234375,
      "learning_rate": 5.186245169149784e-08,
      "loss": 0.5661,
      "mean_token_accuracy": 0.8385386119257087,
      "step": 400
    },
    {
      "epoch": 0.9702850212249848,
      "eval_loss": 0.5721316337585449,
      "eval_mean_token_accuracy": 0.8450646699974657,
      "eval_runtime": 0.5673,
      "eval_samples_per_second": 128.69,
      "eval_steps_per_second": 5.289,
      "step": 400
    },
    {
      "epoch": 0.9824135839902971,
      "grad_norm": 0.90234375,
      "learning_rate": 1.7657716077265075e-08,
      "loss": 0.562,
      "mean_token_accuracy": 0.8391877443792767,
      "step": 405
    },
    {
      "epoch": 0.9945421467556095,
      "grad_norm": 0.90234375,
      "learning_rate": 1.4418359715351327e-09,
      "loss": 0.558,
      "mean_token_accuracy": 0.8401652614858259,
      "step": 410
    },
    {
      "epoch": 0.9993935718617344,
      "mean_token_accuracy": 0.8415696786412512,
      "step": 412,
      "total_flos": 1.1596255879128678e+17,
      "train_loss": 0.571030998982272,
      "train_runtime": 553.8515,
      "train_samples_per_second": 95.222,
      "train_steps_per_second": 0.744
    }
  ],
  "logging_steps": 5,
  "max_steps": 412,
  "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": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.1596255879128678e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}