{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.010608828903164967,
  "eval_steps": 500,
  "global_step": 22500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00023575175340366595,
      "grad_norm": 0.8459708094596863,
      "learning_rate": 0.0002999764248246596,
      "loss": 8.8499,
      "step": 500
    },
    {
      "epoch": 0.0004715035068073319,
      "grad_norm": 1.0758037567138672,
      "learning_rate": 0.00029995284964931925,
      "loss": 6.2672,
      "step": 1000
    },
    {
      "epoch": 0.0007072552602109978,
      "grad_norm": 0.9614197611808777,
      "learning_rate": 0.0002999292744739789,
      "loss": 5.7862,
      "step": 1500
    },
    {
      "epoch": 0.0009430070136146638,
      "grad_norm": 0.9660577774047852,
      "learning_rate": 0.0002999056992986385,
      "loss": 5.5715,
      "step": 2000
    },
    {
      "epoch": 0.0011787587670183297,
      "grad_norm": 1.3354485034942627,
      "learning_rate": 0.00029988212412329813,
      "loss": 5.4455,
      "step": 2500
    },
    {
      "epoch": 0.0014145105204219955,
      "grad_norm": 1.0552821159362793,
      "learning_rate": 0.00029985854894795774,
      "loss": 5.3649,
      "step": 3000
    },
    {
      "epoch": 0.0016502622738256616,
      "grad_norm": 1.069198489189148,
      "learning_rate": 0.0002998349737726174,
      "loss": 5.281,
      "step": 3500
    },
    {
      "epoch": 0.0018860140272293276,
      "grad_norm": 1.2014654874801636,
      "learning_rate": 0.00029981139859727707,
      "loss": 5.2006,
      "step": 4000
    },
    {
      "epoch": 0.0021217657806329934,
      "grad_norm": 1.0098108053207397,
      "learning_rate": 0.0002997878234219367,
      "loss": 5.1572,
      "step": 4500
    },
    {
      "epoch": 0.0023575175340366595,
      "grad_norm": 1.2696442604064941,
      "learning_rate": 0.0002997642482465963,
      "loss": 5.1337,
      "step": 5000
    },
    {
      "epoch": 0.0025932692874403255,
      "grad_norm": 1.1595019102096558,
      "learning_rate": 0.00029974067307125596,
      "loss": 5.0396,
      "step": 5500
    },
    {
      "epoch": 0.002829021040843991,
      "grad_norm": 1.01584792137146,
      "learning_rate": 0.00029971709789591557,
      "loss": 5.0324,
      "step": 6000
    },
    {
      "epoch": 0.003064772794247657,
      "grad_norm": 1.6431899070739746,
      "learning_rate": 0.00029969352272057523,
      "loss": 4.9911,
      "step": 6500
    },
    {
      "epoch": 0.003300524547651323,
      "grad_norm": 0.9707762002944946,
      "learning_rate": 0.00029966994754523484,
      "loss": 4.9459,
      "step": 7000
    },
    {
      "epoch": 0.003536276301054989,
      "grad_norm": 1.0742356777191162,
      "learning_rate": 0.00029964637236989445,
      "loss": 4.8719,
      "step": 7500
    },
    {
      "epoch": 0.0037720280544586552,
      "grad_norm": 1.077572226524353,
      "learning_rate": 0.0002996227971945541,
      "loss": 4.8526,
      "step": 8000
    },
    {
      "epoch": 0.004007779807862321,
      "grad_norm": 1.2099336385726929,
      "learning_rate": 0.0002995992220192138,
      "loss": 4.805,
      "step": 8500
    },
    {
      "epoch": 0.004243531561265987,
      "grad_norm": 1.2295851707458496,
      "learning_rate": 0.0002995756468438734,
      "loss": 4.7844,
      "step": 9000
    },
    {
      "epoch": 0.004479283314669653,
      "grad_norm": 1.419725775718689,
      "learning_rate": 0.000299552071668533,
      "loss": 4.7182,
      "step": 9500
    },
    {
      "epoch": 0.004715035068073319,
      "grad_norm": 1.2460483312606812,
      "learning_rate": 0.00029952849649319266,
      "loss": 4.7153,
      "step": 10000
    },
    {
      "epoch": 0.004950786821476985,
      "grad_norm": 1.3061468601226807,
      "learning_rate": 0.0002995049213178523,
      "loss": 4.6789,
      "step": 10500
    },
    {
      "epoch": 0.005186538574880651,
      "grad_norm": 1.0660468339920044,
      "learning_rate": 0.00029948134614251194,
      "loss": 4.6484,
      "step": 11000
    },
    {
      "epoch": 0.005422290328284317,
      "grad_norm": 1.0721254348754883,
      "learning_rate": 0.00029945777096717155,
      "loss": 4.6284,
      "step": 11500
    },
    {
      "epoch": 0.005658042081687982,
      "grad_norm": 1.1215749979019165,
      "learning_rate": 0.00029943419579183116,
      "loss": 4.5578,
      "step": 12000
    },
    {
      "epoch": 0.005893793835091648,
      "grad_norm": 0.9331501126289368,
      "learning_rate": 0.0002994106206164908,
      "loss": 4.5434,
      "step": 12500
    },
    {
      "epoch": 0.006129545588495314,
      "grad_norm": 1.6719545125961304,
      "learning_rate": 0.00029938704544115043,
      "loss": 4.5433,
      "step": 13000
    },
    {
      "epoch": 0.00636529734189898,
      "grad_norm": 1.087511658668518,
      "learning_rate": 0.0002993634702658101,
      "loss": 4.451,
      "step": 13500
    },
    {
      "epoch": 0.006601049095302646,
      "grad_norm": 0.9610065817832947,
      "learning_rate": 0.0002993398950904697,
      "loss": 4.4753,
      "step": 14000
    },
    {
      "epoch": 0.006836800848706312,
      "grad_norm": 1.6184645891189575,
      "learning_rate": 0.0002993163199151293,
      "loss": 4.4395,
      "step": 14500
    },
    {
      "epoch": 0.007072552602109978,
      "grad_norm": 1.266706109046936,
      "learning_rate": 0.000299292744739789,
      "loss": 4.4283,
      "step": 15000
    },
    {
      "epoch": 0.007308304355513644,
      "grad_norm": 1.0746177434921265,
      "learning_rate": 0.0002992691695644486,
      "loss": 4.3878,
      "step": 15500
    },
    {
      "epoch": 0.0075440561089173104,
      "grad_norm": 1.0867644548416138,
      "learning_rate": 0.00029924559438910826,
      "loss": 4.3745,
      "step": 16000
    },
    {
      "epoch": 0.007779807862320976,
      "grad_norm": 1.246843934059143,
      "learning_rate": 0.00029922201921376787,
      "loss": 4.3288,
      "step": 16500
    },
    {
      "epoch": 0.008015559615724642,
      "grad_norm": 1.013817310333252,
      "learning_rate": 0.00029919844403842753,
      "loss": 4.3222,
      "step": 17000
    },
    {
      "epoch": 0.008251311369128309,
      "grad_norm": 0.7903661727905273,
      "learning_rate": 0.00029917486886308714,
      "loss": 4.2851,
      "step": 17500
    },
    {
      "epoch": 0.008487063122531974,
      "grad_norm": 0.884263277053833,
      "learning_rate": 0.00029915129368774675,
      "loss": 4.2747,
      "step": 18000
    },
    {
      "epoch": 0.00872281487593564,
      "grad_norm": 0.901438295841217,
      "learning_rate": 0.0002991277185124064,
      "loss": 4.248,
      "step": 18500
    },
    {
      "epoch": 0.008958566629339306,
      "grad_norm": 1.007119059562683,
      "learning_rate": 0.000299104143337066,
      "loss": 4.2322,
      "step": 19000
    },
    {
      "epoch": 0.009194318382742971,
      "grad_norm": 0.9175025224685669,
      "learning_rate": 0.0002990805681617257,
      "loss": 4.224,
      "step": 19500
    },
    {
      "epoch": 0.009430070136146638,
      "grad_norm": 0.802945077419281,
      "learning_rate": 0.0002990569929863853,
      "loss": 4.1627,
      "step": 20000
    },
    {
      "epoch": 0.009665821889550303,
      "grad_norm": 0.9863154292106628,
      "learning_rate": 0.0002990334178110449,
      "loss": 4.159,
      "step": 20500
    },
    {
      "epoch": 0.00990157364295397,
      "grad_norm": 0.9913619160652161,
      "learning_rate": 0.0002990098426357046,
      "loss": 4.1331,
      "step": 21000
    },
    {
      "epoch": 0.010137325396357635,
      "grad_norm": 0.9557477831840515,
      "learning_rate": 0.00029898626746036424,
      "loss": 4.1369,
      "step": 21500
    },
    {
      "epoch": 0.010373077149761302,
      "grad_norm": 0.9752131104469299,
      "learning_rate": 0.00029896269228502385,
      "loss": 4.1304,
      "step": 22000
    },
    {
      "epoch": 0.010608828903164967,
      "grad_norm": 0.8786485195159912,
      "learning_rate": 0.00029893911710968346,
      "loss": 4.1096,
      "step": 22500
    }
  ],
  "logging_steps": 500,
  "max_steps": 6362625,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 2.4361403726168064e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}