{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9071323279283365,
  "eval_steps": 5000,
  "global_step": 40000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.011339154099104206,
      "grad_norm": 0.8725878596305847,
      "learning_rate": 4.943304229504479e-05,
      "loss": 1.6299,
      "step": 500
    },
    {
      "epoch": 0.022678308198208413,
      "grad_norm": 0.6967089176177979,
      "learning_rate": 4.886608459008958e-05,
      "loss": 1.5715,
      "step": 1000
    },
    {
      "epoch": 0.03401746229731262,
      "grad_norm": 1.339996099472046,
      "learning_rate": 4.830026080054428e-05,
      "loss": 1.5027,
      "step": 1500
    },
    {
      "epoch": 0.045356616396416825,
      "grad_norm": 1.1787469387054443,
      "learning_rate": 4.773330309558907e-05,
      "loss": 1.5426,
      "step": 2000
    },
    {
      "epoch": 0.05669577049552103,
      "grad_norm": 1.0006368160247803,
      "learning_rate": 4.716634539063386e-05,
      "loss": 1.4986,
      "step": 2500
    },
    {
      "epoch": 0.06803492459462523,
      "grad_norm": 2.14919114112854,
      "learning_rate": 4.659938768567865e-05,
      "loss": 1.5024,
      "step": 3000
    },
    {
      "epoch": 0.07937407869372945,
      "grad_norm": 1.070865511894226,
      "learning_rate": 4.603242998072344e-05,
      "loss": 1.4799,
      "step": 3500
    },
    {
      "epoch": 0.09071323279283365,
      "grad_norm": 2.105903387069702,
      "learning_rate": 4.546547227576823e-05,
      "loss": 1.4478,
      "step": 4000
    },
    {
      "epoch": 0.10205238689193787,
      "grad_norm": 1.5750607252120972,
      "learning_rate": 4.489851457081302e-05,
      "loss": 1.4962,
      "step": 4500
    },
    {
      "epoch": 0.11339154099104207,
      "grad_norm": 1.0978702306747437,
      "learning_rate": 4.433269078126772e-05,
      "loss": 1.4764,
      "step": 5000
    },
    {
      "epoch": 0.11339154099104207,
      "eval_loss": 1.4744130373001099,
      "eval_runtime": 416.1836,
      "eval_samples_per_second": 11.774,
      "eval_steps_per_second": 11.774,
      "step": 5000
    },
    {
      "epoch": 0.12473069509014628,
      "grad_norm": 1.209443211555481,
      "learning_rate": 4.376573307631251e-05,
      "loss": 1.3968,
      "step": 5500
    },
    {
      "epoch": 0.13606984918925047,
      "grad_norm": 1.441433310508728,
      "learning_rate": 4.31987753713573e-05,
      "loss": 1.3486,
      "step": 6000
    },
    {
      "epoch": 0.14740900328835468,
      "grad_norm": 1.2752562761306763,
      "learning_rate": 4.263181766640209e-05,
      "loss": 1.4045,
      "step": 6500
    },
    {
      "epoch": 0.1587481573874589,
      "grad_norm": 1.398257851600647,
      "learning_rate": 4.206485996144688e-05,
      "loss": 1.3973,
      "step": 7000
    },
    {
      "epoch": 0.17008731148656311,
      "grad_norm": 1.952150821685791,
      "learning_rate": 4.149903617190158e-05,
      "loss": 1.4073,
      "step": 7500
    },
    {
      "epoch": 0.1814264655856673,
      "grad_norm": 1.4155147075653076,
      "learning_rate": 4.093207846694637e-05,
      "loss": 1.4652,
      "step": 8000
    },
    {
      "epoch": 0.19276561968477152,
      "grad_norm": 0.8617384433746338,
      "learning_rate": 4.036512076199116e-05,
      "loss": 1.4007,
      "step": 8500
    },
    {
      "epoch": 0.20410477378387573,
      "grad_norm": 1.318098783493042,
      "learning_rate": 3.9798163057035946e-05,
      "loss": 1.4018,
      "step": 9000
    },
    {
      "epoch": 0.21544392788297992,
      "grad_norm": 1.1673972606658936,
      "learning_rate": 3.9231205352080735e-05,
      "loss": 1.3789,
      "step": 9500
    },
    {
      "epoch": 0.22678308198208413,
      "grad_norm": 1.6874651908874512,
      "learning_rate": 3.866424764712553e-05,
      "loss": 1.3834,
      "step": 10000
    },
    {
      "epoch": 0.22678308198208413,
      "eval_loss": 1.4033006429672241,
      "eval_runtime": 416.1008,
      "eval_samples_per_second": 11.776,
      "eval_steps_per_second": 11.776,
      "step": 10000
    },
    {
      "epoch": 0.23812223608118835,
      "grad_norm": 1.585263967514038,
      "learning_rate": 3.8097289942170314e-05,
      "loss": 1.3716,
      "step": 10500
    },
    {
      "epoch": 0.24946139018029256,
      "grad_norm": 1.8377569913864136,
      "learning_rate": 3.7531466152625015e-05,
      "loss": 1.338,
      "step": 11000
    },
    {
      "epoch": 0.26080054427939675,
      "grad_norm": 1.0479376316070557,
      "learning_rate": 3.6964508447669805e-05,
      "loss": 1.3929,
      "step": 11500
    },
    {
      "epoch": 0.27213969837850094,
      "grad_norm": 1.2484978437423706,
      "learning_rate": 3.63975507427146e-05,
      "loss": 1.3695,
      "step": 12000
    },
    {
      "epoch": 0.2834788524776052,
      "grad_norm": 0.8838484883308411,
      "learning_rate": 3.583059303775938e-05,
      "loss": 1.3883,
      "step": 12500
    },
    {
      "epoch": 0.29481800657670937,
      "grad_norm": 4.149050235748291,
      "learning_rate": 3.526363533280417e-05,
      "loss": 1.354,
      "step": 13000
    },
    {
      "epoch": 0.3061571606758136,
      "grad_norm": 1.2994155883789062,
      "learning_rate": 3.469667762784896e-05,
      "loss": 1.3775,
      "step": 13500
    },
    {
      "epoch": 0.3174963147749178,
      "grad_norm": 0.8917070627212524,
      "learning_rate": 3.412971992289375e-05,
      "loss": 1.2917,
      "step": 14000
    },
    {
      "epoch": 0.328835468874022,
      "grad_norm": 1.4100022315979004,
      "learning_rate": 3.356276221793854e-05,
      "loss": 1.3751,
      "step": 14500
    },
    {
      "epoch": 0.34017462297312623,
      "grad_norm": 1.0797368288040161,
      "learning_rate": 3.299580451298333e-05,
      "loss": 1.3222,
      "step": 15000
    },
    {
      "epoch": 0.34017462297312623,
      "eval_loss": 1.3495228290557861,
      "eval_runtime": 416.0029,
      "eval_samples_per_second": 11.779,
      "eval_steps_per_second": 11.779,
      "step": 15000
    },
    {
      "epoch": 0.3515137770722304,
      "grad_norm": 1.4838085174560547,
      "learning_rate": 3.242998072343803e-05,
      "loss": 1.2672,
      "step": 15500
    },
    {
      "epoch": 0.3628529311713346,
      "grad_norm": 1.1724998950958252,
      "learning_rate": 3.186302301848282e-05,
      "loss": 1.3398,
      "step": 16000
    },
    {
      "epoch": 0.37419208527043885,
      "grad_norm": 1.8046358823776245,
      "learning_rate": 3.129719922893752e-05,
      "loss": 1.3214,
      "step": 16500
    },
    {
      "epoch": 0.38553123936954303,
      "grad_norm": 3.12375545501709,
      "learning_rate": 3.073024152398231e-05,
      "loss": 1.3307,
      "step": 17000
    },
    {
      "epoch": 0.3968703934686472,
      "grad_norm": 5.761733531951904,
      "learning_rate": 3.01632838190271e-05,
      "loss": 1.2611,
      "step": 17500
    },
    {
      "epoch": 0.40820954756775146,
      "grad_norm": 3.4179553985595703,
      "learning_rate": 2.9596326114071893e-05,
      "loss": 1.2736,
      "step": 18000
    },
    {
      "epoch": 0.41954870166685565,
      "grad_norm": 1.7887245416641235,
      "learning_rate": 2.902936840911668e-05,
      "loss": 1.3582,
      "step": 18500
    },
    {
      "epoch": 0.43088785576595984,
      "grad_norm": 1.874280571937561,
      "learning_rate": 2.846241070416147e-05,
      "loss": 1.3569,
      "step": 19000
    },
    {
      "epoch": 0.4422270098650641,
      "grad_norm": 0.8716740608215332,
      "learning_rate": 2.789545299920626e-05,
      "loss": 1.2819,
      "step": 19500
    },
    {
      "epoch": 0.45356616396416827,
      "grad_norm": 4.057077407836914,
      "learning_rate": 2.732849529425105e-05,
      "loss": 1.2754,
      "step": 20000
    },
    {
      "epoch": 0.45356616396416827,
      "eval_loss": 1.3207719326019287,
      "eval_runtime": 416.1264,
      "eval_samples_per_second": 11.775,
      "eval_steps_per_second": 11.775,
      "step": 20000
    },
    {
      "epoch": 0.46490531806327245,
      "grad_norm": 0.8342803716659546,
      "learning_rate": 2.6762671504705748e-05,
      "loss": 1.2799,
      "step": 20500
    },
    {
      "epoch": 0.4762444721623767,
      "grad_norm": 1.0279159545898438,
      "learning_rate": 2.6195713799750538e-05,
      "loss": 1.3015,
      "step": 21000
    },
    {
      "epoch": 0.4875836262614809,
      "grad_norm": 2.64371395111084,
      "learning_rate": 2.562989001020524e-05,
      "loss": 1.3406,
      "step": 21500
    },
    {
      "epoch": 0.4989227803605851,
      "grad_norm": 2.1410977840423584,
      "learning_rate": 2.506293230525003e-05,
      "loss": 1.2915,
      "step": 22000
    },
    {
      "epoch": 0.5102619344596893,
      "grad_norm": 1.1181799173355103,
      "learning_rate": 2.4495974600294817e-05,
      "loss": 1.2353,
      "step": 22500
    },
    {
      "epoch": 0.5216010885587935,
      "grad_norm": 1.3449642658233643,
      "learning_rate": 2.392901689533961e-05,
      "loss": 1.3209,
      "step": 23000
    },
    {
      "epoch": 0.5329402426578977,
      "grad_norm": 12.50430965423584,
      "learning_rate": 2.3362059190384396e-05,
      "loss": 1.2993,
      "step": 23500
    },
    {
      "epoch": 0.5442793967570019,
      "grad_norm": 1.8638157844543457,
      "learning_rate": 2.279510148542919e-05,
      "loss": 1.2416,
      "step": 24000
    },
    {
      "epoch": 0.5556185508561061,
      "grad_norm": 1.905992865562439,
      "learning_rate": 2.2228143780473978e-05,
      "loss": 1.291,
      "step": 24500
    },
    {
      "epoch": 0.5669577049552104,
      "grad_norm": 1.2208611965179443,
      "learning_rate": 2.166231999092868e-05,
      "loss": 1.2926,
      "step": 25000
    },
    {
      "epoch": 0.5669577049552104,
      "eval_loss": 1.3028029203414917,
      "eval_runtime": 415.3527,
      "eval_samples_per_second": 11.797,
      "eval_steps_per_second": 11.797,
      "step": 25000
    },
    {
      "epoch": 0.5782968590543145,
      "grad_norm": 2.476512908935547,
      "learning_rate": 2.1095362285973465e-05,
      "loss": 1.2751,
      "step": 25500
    },
    {
      "epoch": 0.5896360131534187,
      "grad_norm": 1.8113676309585571,
      "learning_rate": 2.0528404581018258e-05,
      "loss": 1.3305,
      "step": 26000
    },
    {
      "epoch": 0.600975167252523,
      "grad_norm": 2.3951265811920166,
      "learning_rate": 1.9961446876063047e-05,
      "loss": 1.3198,
      "step": 26500
    },
    {
      "epoch": 0.6123143213516272,
      "grad_norm": 5.352144718170166,
      "learning_rate": 1.9394489171107837e-05,
      "loss": 1.2682,
      "step": 27000
    },
    {
      "epoch": 0.6236534754507314,
      "grad_norm": 1.3964953422546387,
      "learning_rate": 1.8827531466152626e-05,
      "loss": 1.2538,
      "step": 27500
    },
    {
      "epoch": 0.6349926295498356,
      "grad_norm": 2.935234308242798,
      "learning_rate": 1.8260573761197416e-05,
      "loss": 1.3016,
      "step": 28000
    },
    {
      "epoch": 0.6463317836489398,
      "grad_norm": 1.433309555053711,
      "learning_rate": 1.7693616056242205e-05,
      "loss": 1.3036,
      "step": 28500
    },
    {
      "epoch": 0.657670937748044,
      "grad_norm": 1.0000214576721191,
      "learning_rate": 1.7127792266696906e-05,
      "loss": 1.294,
      "step": 29000
    },
    {
      "epoch": 0.6690100918471482,
      "grad_norm": 1.4275603294372559,
      "learning_rate": 1.6560834561741695e-05,
      "loss": 1.321,
      "step": 29500
    },
    {
      "epoch": 0.6803492459462525,
      "grad_norm": 0.913606584072113,
      "learning_rate": 1.5993876856786485e-05,
      "loss": 1.2988,
      "step": 30000
    },
    {
      "epoch": 0.6803492459462525,
      "eval_loss": 1.2923102378845215,
      "eval_runtime": 415.3403,
      "eval_samples_per_second": 11.798,
      "eval_steps_per_second": 11.798,
      "step": 30000
    },
    {
      "epoch": 0.6916884000453566,
      "grad_norm": 1.5725594758987427,
      "learning_rate": 1.5426919151831274e-05,
      "loss": 1.2599,
      "step": 30500
    },
    {
      "epoch": 0.7030275541444608,
      "grad_norm": 0.9978646039962769,
      "learning_rate": 1.4859961446876063e-05,
      "loss": 1.2287,
      "step": 31000
    },
    {
      "epoch": 0.7143667082435651,
      "grad_norm": 0.7757058143615723,
      "learning_rate": 1.4294137657330764e-05,
      "loss": 1.1999,
      "step": 31500
    },
    {
      "epoch": 0.7257058623426692,
      "grad_norm": 0.9467388391494751,
      "learning_rate": 1.3727179952375554e-05,
      "loss": 1.3419,
      "step": 32000
    },
    {
      "epoch": 0.7370450164417734,
      "grad_norm": 0.8585200309753418,
      "learning_rate": 1.3161356162830255e-05,
      "loss": 1.2497,
      "step": 32500
    },
    {
      "epoch": 0.7483841705408777,
      "grad_norm": 1.6182235479354858,
      "learning_rate": 1.2594398457875043e-05,
      "loss": 1.2516,
      "step": 33000
    },
    {
      "epoch": 0.7597233246399818,
      "grad_norm": 0.916101336479187,
      "learning_rate": 1.2027440752919834e-05,
      "loss": 1.2583,
      "step": 33500
    },
    {
      "epoch": 0.7710624787390861,
      "grad_norm": 2.820047616958618,
      "learning_rate": 1.1460483047964623e-05,
      "loss": 1.304,
      "step": 34000
    },
    {
      "epoch": 0.7824016328381903,
      "grad_norm": 1.3022847175598145,
      "learning_rate": 1.089352534300941e-05,
      "loss": 1.1609,
      "step": 34500
    },
    {
      "epoch": 0.7937407869372944,
      "grad_norm": 1.2741048336029053,
      "learning_rate": 1.0326567638054202e-05,
      "loss": 1.2231,
      "step": 35000
    },
    {
      "epoch": 0.7937407869372944,
      "eval_loss": 1.2846976518630981,
      "eval_runtime": 415.5398,
      "eval_samples_per_second": 11.792,
      "eval_steps_per_second": 11.792,
      "step": 35000
    },
    {
      "epoch": 0.8050799410363987,
      "grad_norm": 1.3110229969024658,
      "learning_rate": 9.759609933098991e-06,
      "loss": 1.2554,
      "step": 35500
    },
    {
      "epoch": 0.8164190951355029,
      "grad_norm": 2.569902181625366,
      "learning_rate": 9.19265222814378e-06,
      "loss": 1.2252,
      "step": 36000
    },
    {
      "epoch": 0.8277582492346071,
      "grad_norm": 1.4749958515167236,
      "learning_rate": 8.62682843859848e-06,
      "loss": 1.2405,
      "step": 36500
    },
    {
      "epoch": 0.8390974033337113,
      "grad_norm": 1.2822141647338867,
      "learning_rate": 8.059870733643271e-06,
      "loss": 1.241,
      "step": 37000
    },
    {
      "epoch": 0.8504365574328155,
      "grad_norm": 1.817275881767273,
      "learning_rate": 7.49291302868806e-06,
      "loss": 1.2054,
      "step": 37500
    },
    {
      "epoch": 0.8617757115319197,
      "grad_norm": 1.796393632888794,
      "learning_rate": 6.92595532373285e-06,
      "loss": 1.2011,
      "step": 38000
    },
    {
      "epoch": 0.8731148656310239,
      "grad_norm": 1.1779612302780151,
      "learning_rate": 6.35899761877764e-06,
      "loss": 1.2521,
      "step": 38500
    },
    {
      "epoch": 0.8844540197301282,
      "grad_norm": 1.9495667219161987,
      "learning_rate": 5.792039913822429e-06,
      "loss": 1.2747,
      "step": 39000
    },
    {
      "epoch": 0.8957931738292323,
      "grad_norm": 1.5888334512710571,
      "learning_rate": 5.226216124277129e-06,
      "loss": 1.2255,
      "step": 39500
    },
    {
      "epoch": 0.9071323279283365,
      "grad_norm": 1.0815423727035522,
      "learning_rate": 4.659258419321919e-06,
      "loss": 1.2197,
      "step": 40000
    },
    {
      "epoch": 0.9071323279283365,
      "eval_loss": 1.2798832654953003,
      "eval_runtime": 415.0622,
      "eval_samples_per_second": 11.805,
      "eval_steps_per_second": 11.805,
      "step": 40000
    }
  ],
  "logging_steps": 500,
  "max_steps": 44095,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.1124278259728384e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}