{
  "best_metric": 2.8960378170013428,
  "best_model_checkpoint": "outputs-6_7/checkpoint-64000",
  "epoch": 2.7480548923964756,
  "eval_steps": 4000,
  "global_step": 64000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.021469178846847466,
      "grad_norm": 0.39146578311920166,
      "learning_rate": 0.00029989693820586593,
      "loss": 2.4186,
      "step": 500
    },
    {
      "epoch": 0.04293835769369493,
      "grad_norm": 0.5122425556182861,
      "learning_rate": 0.0002997681109631983,
      "loss": 2.3709,
      "step": 1000
    },
    {
      "epoch": 0.0644075365405424,
      "grad_norm": 0.4298762083053589,
      "learning_rate": 0.0002996392837205307,
      "loss": 2.3735,
      "step": 1500
    },
    {
      "epoch": 0.08587671538738986,
      "grad_norm": 0.39066824316978455,
      "learning_rate": 0.00029951045647786317,
      "loss": 2.3524,
      "step": 2000
    },
    {
      "epoch": 0.10734589423423734,
      "grad_norm": 0.39771586656570435,
      "learning_rate": 0.00029938162923519556,
      "loss": 2.347,
      "step": 2500
    },
    {
      "epoch": 0.1288150730810848,
      "grad_norm": 0.47568196058273315,
      "learning_rate": 0.000299252801992528,
      "loss": 2.3336,
      "step": 3000
    },
    {
      "epoch": 0.15028425192793227,
      "grad_norm": 0.44162309169769287,
      "learning_rate": 0.0002991239747498604,
      "loss": 2.334,
      "step": 3500
    },
    {
      "epoch": 0.17175343077477973,
      "grad_norm": 0.4031461179256439,
      "learning_rate": 0.0002989951475071928,
      "loss": 2.3266,
      "step": 4000
    },
    {
      "epoch": 0.17175343077477973,
      "eval_loss": 3.0375378131866455,
      "eval_runtime": 174.8485,
      "eval_samples_per_second": 14.298,
      "eval_steps_per_second": 3.575,
      "step": 4000
    },
    {
      "epoch": 0.19322260962162718,
      "grad_norm": 0.384034126996994,
      "learning_rate": 0.00029886632026452525,
      "loss": 2.3344,
      "step": 4500
    },
    {
      "epoch": 0.21469178846847467,
      "grad_norm": 0.44177886843681335,
      "learning_rate": 0.00029873749302185764,
      "loss": 2.3157,
      "step": 5000
    },
    {
      "epoch": 0.23616096731532213,
      "grad_norm": 0.4425281286239624,
      "learning_rate": 0.0002986086657791901,
      "loss": 2.3232,
      "step": 5500
    },
    {
      "epoch": 0.2576301461621696,
      "grad_norm": 0.4302816390991211,
      "learning_rate": 0.0002984798385365225,
      "loss": 2.3141,
      "step": 6000
    },
    {
      "epoch": 0.2790993250090171,
      "grad_norm": 0.5806054472923279,
      "learning_rate": 0.00029835101129385493,
      "loss": 2.2975,
      "step": 6500
    },
    {
      "epoch": 0.30056850385586453,
      "grad_norm": 0.5654121041297913,
      "learning_rate": 0.00029822218405118733,
      "loss": 2.3141,
      "step": 7000
    },
    {
      "epoch": 0.322037682702712,
      "grad_norm": 0.5454065203666687,
      "learning_rate": 0.0002980933568085197,
      "loss": 2.3097,
      "step": 7500
    },
    {
      "epoch": 0.34350686154955945,
      "grad_norm": 0.43060022592544556,
      "learning_rate": 0.00029796452956585217,
      "loss": 2.308,
      "step": 8000
    },
    {
      "epoch": 0.34350686154955945,
      "eval_loss": 3.0060064792633057,
      "eval_runtime": 171.3325,
      "eval_samples_per_second": 14.592,
      "eval_steps_per_second": 3.648,
      "step": 8000
    },
    {
      "epoch": 0.3649760403964069,
      "grad_norm": 0.490461140871048,
      "learning_rate": 0.00029783570232318456,
      "loss": 2.2985,
      "step": 8500
    },
    {
      "epoch": 0.38644521924325437,
      "grad_norm": 0.5096587538719177,
      "learning_rate": 0.000297706875080517,
      "loss": 2.3041,
      "step": 9000
    },
    {
      "epoch": 0.4079143980901018,
      "grad_norm": 0.4906415343284607,
      "learning_rate": 0.0002975780478378494,
      "loss": 2.2903,
      "step": 9500
    },
    {
      "epoch": 0.42938357693694934,
      "grad_norm": 0.5885447263717651,
      "learning_rate": 0.00029744922059518186,
      "loss": 2.3069,
      "step": 10000
    },
    {
      "epoch": 0.4508527557837968,
      "grad_norm": 0.5200388431549072,
      "learning_rate": 0.00029732039335251425,
      "loss": 2.3025,
      "step": 10500
    },
    {
      "epoch": 0.47232193463064426,
      "grad_norm": 0.6331049799919128,
      "learning_rate": 0.00029719156610984664,
      "loss": 2.2957,
      "step": 11000
    },
    {
      "epoch": 0.4937911134774917,
      "grad_norm": 0.5442560315132141,
      "learning_rate": 0.0002970627388671791,
      "loss": 2.2878,
      "step": 11500
    },
    {
      "epoch": 0.5152602923243392,
      "grad_norm": 0.5305426120758057,
      "learning_rate": 0.0002969339116245115,
      "loss": 2.2903,
      "step": 12000
    },
    {
      "epoch": 0.5152602923243392,
      "eval_loss": 2.973823070526123,
      "eval_runtime": 170.1319,
      "eval_samples_per_second": 14.694,
      "eval_steps_per_second": 3.674,
      "step": 12000
    },
    {
      "epoch": 0.5367294711711866,
      "grad_norm": 0.5756106972694397,
      "learning_rate": 0.0002968050843818439,
      "loss": 2.2883,
      "step": 12500
    },
    {
      "epoch": 0.5581986500180341,
      "grad_norm": 0.5812390446662903,
      "learning_rate": 0.00029667625713917633,
      "loss": 2.2807,
      "step": 13000
    },
    {
      "epoch": 0.5796678288648816,
      "grad_norm": 0.4355560541152954,
      "learning_rate": 0.0002965474298965088,
      "loss": 2.2885,
      "step": 13500
    },
    {
      "epoch": 0.6011370077117291,
      "grad_norm": 0.41715824604034424,
      "learning_rate": 0.00029641860265384117,
      "loss": 2.2834,
      "step": 14000
    },
    {
      "epoch": 0.6226061865585765,
      "grad_norm": 0.4623817801475525,
      "learning_rate": 0.00029628977541117357,
      "loss": 2.2748,
      "step": 14500
    },
    {
      "epoch": 0.644075365405424,
      "grad_norm": 0.5191289186477661,
      "learning_rate": 0.000296160948168506,
      "loss": 2.2811,
      "step": 15000
    },
    {
      "epoch": 0.6655445442522714,
      "grad_norm": 0.6877865791320801,
      "learning_rate": 0.0002960321209258384,
      "loss": 2.2783,
      "step": 15500
    },
    {
      "epoch": 0.6870137230991189,
      "grad_norm": 0.49987566471099854,
      "learning_rate": 0.0002959032936831708,
      "loss": 2.2719,
      "step": 16000
    },
    {
      "epoch": 0.6870137230991189,
      "eval_loss": 2.964353561401367,
      "eval_runtime": 171.8492,
      "eval_samples_per_second": 14.548,
      "eval_steps_per_second": 3.637,
      "step": 16000
    },
    {
      "epoch": 0.7084829019459664,
      "grad_norm": 0.5470739006996155,
      "learning_rate": 0.00029577446644050325,
      "loss": 2.2832,
      "step": 16500
    },
    {
      "epoch": 0.7299520807928138,
      "grad_norm": 0.6002724766731262,
      "learning_rate": 0.0002956456391978357,
      "loss": 2.2838,
      "step": 17000
    },
    {
      "epoch": 0.7514212596396613,
      "grad_norm": 0.6674920320510864,
      "learning_rate": 0.0002955168119551681,
      "loss": 2.2686,
      "step": 17500
    },
    {
      "epoch": 0.7728904384865087,
      "grad_norm": 0.5728652477264404,
      "learning_rate": 0.0002953879847125005,
      "loss": 2.2725,
      "step": 18000
    },
    {
      "epoch": 0.7943596173333562,
      "grad_norm": 0.5590266585350037,
      "learning_rate": 0.00029525915746983294,
      "loss": 2.2794,
      "step": 18500
    },
    {
      "epoch": 0.8158287961802037,
      "grad_norm": 0.7446316480636597,
      "learning_rate": 0.00029513033022716533,
      "loss": 2.2676,
      "step": 19000
    },
    {
      "epoch": 0.8372979750270512,
      "grad_norm": 0.4322523772716522,
      "learning_rate": 0.0002950015029844977,
      "loss": 2.2832,
      "step": 19500
    },
    {
      "epoch": 0.8587671538738987,
      "grad_norm": 0.6566835045814514,
      "learning_rate": 0.00029487267574183017,
      "loss": 2.2636,
      "step": 20000
    },
    {
      "epoch": 0.8587671538738987,
      "eval_loss": 2.954716920852661,
      "eval_runtime": 171.5581,
      "eval_samples_per_second": 14.572,
      "eval_steps_per_second": 3.643,
      "step": 20000
    },
    {
      "epoch": 0.8802363327207461,
      "grad_norm": 0.5313192009925842,
      "learning_rate": 0.0002947438484991626,
      "loss": 2.2819,
      "step": 20500
    },
    {
      "epoch": 0.9017055115675936,
      "grad_norm": 0.689608633518219,
      "learning_rate": 0.000294615021256495,
      "loss": 2.2728,
      "step": 21000
    },
    {
      "epoch": 0.923174690414441,
      "grad_norm": 0.7024255394935608,
      "learning_rate": 0.0002944861940138274,
      "loss": 2.2746,
      "step": 21500
    },
    {
      "epoch": 0.9446438692612885,
      "grad_norm": 0.6012333035469055,
      "learning_rate": 0.00029435736677115986,
      "loss": 2.2658,
      "step": 22000
    },
    {
      "epoch": 0.9661130481081359,
      "grad_norm": 0.6304742693901062,
      "learning_rate": 0.0002942285395284923,
      "loss": 2.2718,
      "step": 22500
    },
    {
      "epoch": 0.9875822269549834,
      "grad_norm": 0.541362464427948,
      "learning_rate": 0.0002940997122858247,
      "loss": 2.272,
      "step": 23000
    },
    {
      "epoch": 1.009051405801831,
      "grad_norm": 0.5888085961341858,
      "learning_rate": 0.0002939708850431571,
      "loss": 2.2485,
      "step": 23500
    },
    {
      "epoch": 1.0305205846486785,
      "grad_norm": 0.5453173518180847,
      "learning_rate": 0.00029384205780048954,
      "loss": 2.2395,
      "step": 24000
    },
    {
      "epoch": 1.0305205846486785,
      "eval_loss": 2.940072774887085,
      "eval_runtime": 171.7472,
      "eval_samples_per_second": 14.556,
      "eval_steps_per_second": 3.639,
      "step": 24000
    },
    {
      "epoch": 1.0519897634955258,
      "grad_norm": 0.7155711054801941,
      "learning_rate": 0.00029371323055782194,
      "loss": 2.2476,
      "step": 24500
    },
    {
      "epoch": 1.0734589423423733,
      "grad_norm": 0.7307182550430298,
      "learning_rate": 0.00029358440331515433,
      "loss": 2.2479,
      "step": 25000
    },
    {
      "epoch": 1.0949281211892208,
      "grad_norm": 0.6849473714828491,
      "learning_rate": 0.0002934555760724868,
      "loss": 2.2407,
      "step": 25500
    },
    {
      "epoch": 1.1163973000360683,
      "grad_norm": 0.7161998152732849,
      "learning_rate": 0.00029332674882981923,
      "loss": 2.247,
      "step": 26000
    },
    {
      "epoch": 1.1378664788829158,
      "grad_norm": 0.723235011100769,
      "learning_rate": 0.0002931979215871516,
      "loss": 2.2382,
      "step": 26500
    },
    {
      "epoch": 1.159335657729763,
      "grad_norm": 0.4874274432659149,
      "learning_rate": 0.000293069094344484,
      "loss": 2.2483,
      "step": 27000
    },
    {
      "epoch": 1.1808048365766106,
      "grad_norm": 0.5381557941436768,
      "learning_rate": 0.00029294026710181646,
      "loss": 2.2423,
      "step": 27500
    },
    {
      "epoch": 1.2022740154234581,
      "grad_norm": 0.7897226214408875,
      "learning_rate": 0.00029281143985914886,
      "loss": 2.2538,
      "step": 28000
    },
    {
      "epoch": 1.2022740154234581,
      "eval_loss": 2.926734447479248,
      "eval_runtime": 172.6333,
      "eval_samples_per_second": 14.482,
      "eval_steps_per_second": 3.62,
      "step": 28000
    },
    {
      "epoch": 1.2237431942703056,
      "grad_norm": 0.5494747161865234,
      "learning_rate": 0.00029268261261648125,
      "loss": 2.2441,
      "step": 28500
    },
    {
      "epoch": 1.245212373117153,
      "grad_norm": 0.5955171585083008,
      "learning_rate": 0.0002925537853738137,
      "loss": 2.245,
      "step": 29000
    },
    {
      "epoch": 1.2666815519640005,
      "grad_norm": 0.7213128805160522,
      "learning_rate": 0.0002924249581311461,
      "loss": 2.2488,
      "step": 29500
    },
    {
      "epoch": 1.288150730810848,
      "grad_norm": 0.7488630414009094,
      "learning_rate": 0.00029229613088847854,
      "loss": 2.2412,
      "step": 30000
    },
    {
      "epoch": 1.3096199096576955,
      "grad_norm": 0.5948154330253601,
      "learning_rate": 0.00029216730364581094,
      "loss": 2.2378,
      "step": 30500
    },
    {
      "epoch": 1.3310890885045428,
      "grad_norm": 0.7915855050086975,
      "learning_rate": 0.0002920384764031434,
      "loss": 2.2464,
      "step": 31000
    },
    {
      "epoch": 1.3525582673513903,
      "grad_norm": 0.6043704152107239,
      "learning_rate": 0.0002919096491604758,
      "loss": 2.2421,
      "step": 31500
    },
    {
      "epoch": 1.3740274461982378,
      "grad_norm": 0.5474274158477783,
      "learning_rate": 0.0002917808219178082,
      "loss": 2.2507,
      "step": 32000
    },
    {
      "epoch": 1.3740274461982378,
      "eval_loss": 2.9269840717315674,
      "eval_runtime": 174.9446,
      "eval_samples_per_second": 14.29,
      "eval_steps_per_second": 3.573,
      "step": 32000
    },
    {
      "epoch": 1.3954966250450853,
      "grad_norm": 0.5420586466789246,
      "learning_rate": 0.0002916519946751406,
      "loss": 2.2405,
      "step": 32500
    },
    {
      "epoch": 1.4169658038919328,
      "grad_norm": 0.4751032888889313,
      "learning_rate": 0.000291523167432473,
      "loss": 2.2497,
      "step": 33000
    },
    {
      "epoch": 1.4384349827387801,
      "grad_norm": 0.5793635249137878,
      "learning_rate": 0.00029139434018980547,
      "loss": 2.2448,
      "step": 33500
    },
    {
      "epoch": 1.4599041615856276,
      "grad_norm": 0.6635434031486511,
      "learning_rate": 0.00029126551294713786,
      "loss": 2.2434,
      "step": 34000
    },
    {
      "epoch": 1.4813733404324751,
      "grad_norm": 0.5708619356155396,
      "learning_rate": 0.0002911366857044703,
      "loss": 2.2343,
      "step": 34500
    },
    {
      "epoch": 1.5028425192793224,
      "grad_norm": 0.5989744067192078,
      "learning_rate": 0.0002910078584618027,
      "loss": 2.2388,
      "step": 35000
    },
    {
      "epoch": 1.5243116981261702,
      "grad_norm": 0.746486246585846,
      "learning_rate": 0.0002908790312191351,
      "loss": 2.2484,
      "step": 35500
    },
    {
      "epoch": 1.5457808769730175,
      "grad_norm": 0.6059302687644958,
      "learning_rate": 0.00029075020397646755,
      "loss": 2.2409,
      "step": 36000
    },
    {
      "epoch": 1.5457808769730175,
      "eval_loss": 2.918299913406372,
      "eval_runtime": 171.3628,
      "eval_samples_per_second": 14.589,
      "eval_steps_per_second": 3.647,
      "step": 36000
    },
    {
      "epoch": 1.567250055819865,
      "grad_norm": 0.5767127871513367,
      "learning_rate": 0.00029062137673379994,
      "loss": 2.2459,
      "step": 36500
    },
    {
      "epoch": 1.5887192346667125,
      "grad_norm": 0.6815518736839294,
      "learning_rate": 0.0002904925494911324,
      "loss": 2.2565,
      "step": 37000
    },
    {
      "epoch": 1.6101884135135598,
      "grad_norm": 0.6565374732017517,
      "learning_rate": 0.0002903637222484648,
      "loss": 2.2388,
      "step": 37500
    },
    {
      "epoch": 1.6316575923604075,
      "grad_norm": 0.6622541546821594,
      "learning_rate": 0.0002902348950057972,
      "loss": 2.261,
      "step": 38000
    },
    {
      "epoch": 1.6531267712072548,
      "grad_norm": 0.8162985444068909,
      "learning_rate": 0.0002901060677631296,
      "loss": 2.2495,
      "step": 38500
    },
    {
      "epoch": 1.6745959500541023,
      "grad_norm": 0.5659546852111816,
      "learning_rate": 0.000289977240520462,
      "loss": 2.2385,
      "step": 39000
    },
    {
      "epoch": 1.6960651289009498,
      "grad_norm": 0.5625469088554382,
      "learning_rate": 0.00028984841327779447,
      "loss": 2.2372,
      "step": 39500
    },
    {
      "epoch": 1.7175343077477971,
      "grad_norm": 0.5423092842102051,
      "learning_rate": 0.00028971958603512686,
      "loss": 2.2424,
      "step": 40000
    },
    {
      "epoch": 1.7175343077477971,
      "eval_loss": 2.92061710357666,
      "eval_runtime": 172.1102,
      "eval_samples_per_second": 14.526,
      "eval_steps_per_second": 3.631,
      "step": 40000
    },
    {
      "epoch": 1.7390034865946449,
      "grad_norm": 0.7644880414009094,
      "learning_rate": 0.0002895907587924593,
      "loss": 2.2368,
      "step": 40500
    },
    {
      "epoch": 1.7604726654414922,
      "grad_norm": 0.8192068934440613,
      "learning_rate": 0.0002894619315497917,
      "loss": 2.2357,
      "step": 41000
    },
    {
      "epoch": 1.7819418442883397,
      "grad_norm": 0.6234991550445557,
      "learning_rate": 0.0002893331043071241,
      "loss": 2.2418,
      "step": 41500
    },
    {
      "epoch": 1.8034110231351872,
      "grad_norm": 0.5751623511314392,
      "learning_rate": 0.00028920427706445655,
      "loss": 2.2413,
      "step": 42000
    },
    {
      "epoch": 1.8248802019820345,
      "grad_norm": 0.8999291062355042,
      "learning_rate": 0.00028907544982178894,
      "loss": 2.2356,
      "step": 42500
    },
    {
      "epoch": 1.846349380828882,
      "grad_norm": 0.7696816325187683,
      "learning_rate": 0.00028894662257912133,
      "loss": 2.2427,
      "step": 43000
    },
    {
      "epoch": 1.8678185596757295,
      "grad_norm": 0.6660240292549133,
      "learning_rate": 0.0002888177953364538,
      "loss": 2.2507,
      "step": 43500
    },
    {
      "epoch": 1.889287738522577,
      "grad_norm": 0.6106180548667908,
      "learning_rate": 0.00028868896809378623,
      "loss": 2.2428,
      "step": 44000
    },
    {
      "epoch": 1.889287738522577,
      "eval_loss": 2.9116756916046143,
      "eval_runtime": 172.218,
      "eval_samples_per_second": 14.516,
      "eval_steps_per_second": 3.629,
      "step": 44000
    },
    {
      "epoch": 1.9107569173694245,
      "grad_norm": 0.6366661190986633,
      "learning_rate": 0.0002885601408511186,
      "loss": 2.2408,
      "step": 44500
    },
    {
      "epoch": 1.9322260962162718,
      "grad_norm": 0.7893187403678894,
      "learning_rate": 0.000288431313608451,
      "loss": 2.2369,
      "step": 45000
    },
    {
      "epoch": 1.9536952750631194,
      "grad_norm": 0.633651077747345,
      "learning_rate": 0.00028830248636578347,
      "loss": 2.2431,
      "step": 45500
    },
    {
      "epoch": 1.9751644539099669,
      "grad_norm": 0.7481298446655273,
      "learning_rate": 0.00028817365912311586,
      "loss": 2.2371,
      "step": 46000
    },
    {
      "epoch": 1.9966336327568142,
      "grad_norm": 0.596591055393219,
      "learning_rate": 0.00028804483188044826,
      "loss": 2.2358,
      "step": 46500
    },
    {
      "epoch": 2.018102811603662,
      "grad_norm": 0.7450771927833557,
      "learning_rate": 0.0002879160046377807,
      "loss": 2.2238,
      "step": 47000
    },
    {
      "epoch": 2.039571990450509,
      "grad_norm": 0.6886998414993286,
      "learning_rate": 0.00028778717739511315,
      "loss": 2.218,
      "step": 47500
    },
    {
      "epoch": 2.061041169297357,
      "grad_norm": 0.5555692911148071,
      "learning_rate": 0.00028765835015244555,
      "loss": 2.2151,
      "step": 48000
    },
    {
      "epoch": 2.061041169297357,
      "eval_loss": 2.909609317779541,
      "eval_runtime": 175.1711,
      "eval_samples_per_second": 14.272,
      "eval_steps_per_second": 3.568,
      "step": 48000
    },
    {
      "epoch": 2.082510348144204,
      "grad_norm": 0.7904226183891296,
      "learning_rate": 0.00028752952290977794,
      "loss": 2.2271,
      "step": 48500
    },
    {
      "epoch": 2.1039795269910515,
      "grad_norm": 0.8151206374168396,
      "learning_rate": 0.0002874006956671104,
      "loss": 2.2104,
      "step": 49000
    },
    {
      "epoch": 2.1254487058378992,
      "grad_norm": 0.6942662596702576,
      "learning_rate": 0.0002872718684244428,
      "loss": 2.2197,
      "step": 49500
    },
    {
      "epoch": 2.1469178846847465,
      "grad_norm": 0.6846303939819336,
      "learning_rate": 0.0002871430411817752,
      "loss": 2.2286,
      "step": 50000
    },
    {
      "epoch": 2.1683870635315943,
      "grad_norm": 0.8961315155029297,
      "learning_rate": 0.0002870142139391076,
      "loss": 2.2329,
      "step": 50500
    },
    {
      "epoch": 2.1898562423784416,
      "grad_norm": 0.5635807514190674,
      "learning_rate": 0.0002868853866964401,
      "loss": 2.2216,
      "step": 51000
    },
    {
      "epoch": 2.211325421225289,
      "grad_norm": 0.6454870700836182,
      "learning_rate": 0.00028675655945377247,
      "loss": 2.2119,
      "step": 51500
    },
    {
      "epoch": 2.2327946000721366,
      "grad_norm": 0.5770113468170166,
      "learning_rate": 0.00028662773221110486,
      "loss": 2.2216,
      "step": 52000
    },
    {
      "epoch": 2.2327946000721366,
      "eval_loss": 2.9131178855895996,
      "eval_runtime": 170.3262,
      "eval_samples_per_second": 14.678,
      "eval_steps_per_second": 3.669,
      "step": 52000
    },
    {
      "epoch": 2.254263778918984,
      "grad_norm": 0.6589009165763855,
      "learning_rate": 0.0002864989049684373,
      "loss": 2.2237,
      "step": 52500
    },
    {
      "epoch": 2.2757329577658316,
      "grad_norm": 0.6595714688301086,
      "learning_rate": 0.0002863700777257697,
      "loss": 2.2331,
      "step": 53000
    },
    {
      "epoch": 2.297202136612679,
      "grad_norm": 0.5528385639190674,
      "learning_rate": 0.0002862412504831021,
      "loss": 2.2316,
      "step": 53500
    },
    {
      "epoch": 2.318671315459526,
      "grad_norm": 0.6706179976463318,
      "learning_rate": 0.00028611242324043455,
      "loss": 2.2312,
      "step": 54000
    },
    {
      "epoch": 2.340140494306374,
      "grad_norm": 0.6599323153495789,
      "learning_rate": 0.000285983595997767,
      "loss": 2.2228,
      "step": 54500
    },
    {
      "epoch": 2.3616096731532212,
      "grad_norm": 0.7218915820121765,
      "learning_rate": 0.0002858547687550994,
      "loss": 2.2269,
      "step": 55000
    },
    {
      "epoch": 2.3830788520000685,
      "grad_norm": 0.6501777768135071,
      "learning_rate": 0.0002857259415124318,
      "loss": 2.225,
      "step": 55500
    },
    {
      "epoch": 2.4045480308469163,
      "grad_norm": 0.6774255037307739,
      "learning_rate": 0.00028559711426976423,
      "loss": 2.2193,
      "step": 56000
    },
    {
      "epoch": 2.4045480308469163,
      "eval_loss": 2.90551495552063,
      "eval_runtime": 173.4554,
      "eval_samples_per_second": 14.413,
      "eval_steps_per_second": 3.603,
      "step": 56000
    },
    {
      "epoch": 2.4260172096937636,
      "grad_norm": 0.667073130607605,
      "learning_rate": 0.00028546828702709663,
      "loss": 2.2228,
      "step": 56500
    },
    {
      "epoch": 2.4474863885406113,
      "grad_norm": 0.776077389717102,
      "learning_rate": 0.000285339459784429,
      "loss": 2.226,
      "step": 57000
    },
    {
      "epoch": 2.4689555673874586,
      "grad_norm": 0.7873576879501343,
      "learning_rate": 0.00028521063254176147,
      "loss": 2.2219,
      "step": 57500
    },
    {
      "epoch": 2.490424746234306,
      "grad_norm": 0.7621210813522339,
      "learning_rate": 0.0002850818052990939,
      "loss": 2.2226,
      "step": 58000
    },
    {
      "epoch": 2.5118939250811536,
      "grad_norm": 0.774750828742981,
      "learning_rate": 0.0002849529780564263,
      "loss": 2.2328,
      "step": 58500
    },
    {
      "epoch": 2.533363103928001,
      "grad_norm": 0.707665205001831,
      "learning_rate": 0.0002848241508137587,
      "loss": 2.2342,
      "step": 59000
    },
    {
      "epoch": 2.554832282774848,
      "grad_norm": 0.7524703741073608,
      "learning_rate": 0.00028469532357109116,
      "loss": 2.2296,
      "step": 59500
    },
    {
      "epoch": 2.576301461621696,
      "grad_norm": 0.6186488270759583,
      "learning_rate": 0.00028456649632842355,
      "loss": 2.2282,
      "step": 60000
    },
    {
      "epoch": 2.576301461621696,
      "eval_loss": 2.907226085662842,
      "eval_runtime": 176.429,
      "eval_samples_per_second": 14.17,
      "eval_steps_per_second": 3.543,
      "step": 60000
    },
    {
      "epoch": 2.5977706404685432,
      "grad_norm": 0.6811486482620239,
      "learning_rate": 0.00028443766908575594,
      "loss": 2.2357,
      "step": 60500
    },
    {
      "epoch": 2.619239819315391,
      "grad_norm": 0.7401767373085022,
      "learning_rate": 0.0002843088418430884,
      "loss": 2.2264,
      "step": 61000
    },
    {
      "epoch": 2.6407089981622383,
      "grad_norm": 0.6240813136100769,
      "learning_rate": 0.00028418001460042084,
      "loss": 2.2402,
      "step": 61500
    },
    {
      "epoch": 2.6621781770090855,
      "grad_norm": 0.6217384338378906,
      "learning_rate": 0.00028405118735775324,
      "loss": 2.2294,
      "step": 62000
    },
    {
      "epoch": 2.6836473558559333,
      "grad_norm": 0.5563312768936157,
      "learning_rate": 0.00028392236011508563,
      "loss": 2.2344,
      "step": 62500
    },
    {
      "epoch": 2.7051165347027806,
      "grad_norm": 0.7275550961494446,
      "learning_rate": 0.0002837935328724181,
      "loss": 2.2233,
      "step": 63000
    },
    {
      "epoch": 2.7265857135496283,
      "grad_norm": 0.6657426953315735,
      "learning_rate": 0.00028366470562975047,
      "loss": 2.2269,
      "step": 63500
    },
    {
      "epoch": 2.7480548923964756,
      "grad_norm": 0.5833483934402466,
      "learning_rate": 0.0002835358783870829,
      "loss": 2.2234,
      "step": 64000
    },
    {
      "epoch": 2.7480548923964756,
      "eval_loss": 2.8960378170013428,
      "eval_runtime": 165.7901,
      "eval_samples_per_second": 15.079,
      "eval_steps_per_second": 3.77,
      "step": 64000
    }
  ],
  "logging_steps": 500,
  "max_steps": 1164450,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 8000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 7.740828282336215e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}