{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.3438641736514077,
  "eval_steps": 500,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 0.967700420320034,
      "epoch": 0.008596604341285192,
      "grad_norm": 1.28125,
      "learning_rate": 7.500000000000001e-05,
      "loss": 1.1922,
      "mean_token_accuracy": 0.7462927460670471,
      "num_tokens": 24955.0,
      "step": 10
    },
    {
      "entropy": 0.8809127531945705,
      "epoch": 0.017193208682570384,
      "grad_norm": 0.640625,
      "learning_rate": 0.00015833333333333332,
      "loss": 0.9056,
      "mean_token_accuracy": 0.7825774624943733,
      "num_tokens": 50126.0,
      "step": 20
    },
    {
      "entropy": 0.7974838063120842,
      "epoch": 0.025789813023855575,
      "grad_norm": 0.466796875,
      "learning_rate": 0.0001999795133245889,
      "loss": 0.8262,
      "mean_token_accuracy": 0.7897384226322174,
      "num_tokens": 76211.0,
      "step": 30
    },
    {
      "entropy": 0.6672817915678024,
      "epoch": 0.03438641736514077,
      "grad_norm": 0.53125,
      "learning_rate": 0.00019981567028232514,
      "loss": 0.6561,
      "mean_token_accuracy": 0.8081615135073662,
      "num_tokens": 99676.0,
      "step": 40
    },
    {
      "entropy": 0.7113775327801705,
      "epoch": 0.04298302170642596,
      "grad_norm": 0.41015625,
      "learning_rate": 0.00019948825269822934,
      "loss": 0.6759,
      "mean_token_accuracy": 0.8067922666668892,
      "num_tokens": 122341.0,
      "step": 50
    },
    {
      "entropy": 0.6950222067534924,
      "epoch": 0.05157962604771115,
      "grad_norm": 0.470703125,
      "learning_rate": 0.00019899779713315575,
      "loss": 0.7108,
      "mean_token_accuracy": 0.8030081078410148,
      "num_tokens": 143840.0,
      "step": 60
    },
    {
      "entropy": 0.7753045067191124,
      "epoch": 0.060176230388996346,
      "grad_norm": 0.388671875,
      "learning_rate": 0.00019834510732908315,
      "loss": 0.7512,
      "mean_token_accuracy": 0.7937675878405571,
      "num_tokens": 169630.0,
      "step": 70
    },
    {
      "entropy": 0.7161391645669937,
      "epoch": 0.06877283473028153,
      "grad_norm": 0.384765625,
      "learning_rate": 0.0001975312528919697,
      "loss": 0.7586,
      "mean_token_accuracy": 0.793881094455719,
      "num_tokens": 192514.0,
      "step": 80
    },
    {
      "entropy": 0.7045083746314049,
      "epoch": 0.07736943907156674,
      "grad_norm": 0.353515625,
      "learning_rate": 0.00019655756753891916,
      "loss": 0.6853,
      "mean_token_accuracy": 0.8054596990346908,
      "num_tokens": 212398.0,
      "step": 90
    },
    {
      "entropy": 0.7032747760415077,
      "epoch": 0.08596604341285193,
      "grad_norm": 0.396484375,
      "learning_rate": 0.0001954256469125301,
      "loss": 0.7434,
      "mean_token_accuracy": 0.797465617954731,
      "num_tokens": 236559.0,
      "step": 100
    },
    {
      "entropy": 0.6814083471894264,
      "epoch": 0.09456264775413711,
      "grad_norm": 0.328125,
      "learning_rate": 0.00019413734596601104,
      "loss": 0.653,
      "mean_token_accuracy": 0.809566143155098,
      "num_tokens": 259990.0,
      "step": 110
    },
    {
      "entropy": 0.6919607870280743,
      "epoch": 0.1031592520954223,
      "grad_norm": 0.388671875,
      "learning_rate": 0.0001926947759233459,
      "loss": 0.7015,
      "mean_token_accuracy": 0.8029481008648872,
      "num_tokens": 281771.0,
      "step": 120
    },
    {
      "entropy": 0.7389424860477447,
      "epoch": 0.1117558564367075,
      "grad_norm": 0.353515625,
      "learning_rate": 0.00019110030081949156,
      "loss": 0.7648,
      "mean_token_accuracy": 0.7935838386416435,
      "num_tokens": 305329.0,
      "step": 130
    },
    {
      "entropy": 0.7182212144136428,
      "epoch": 0.12035246077799269,
      "grad_norm": 0.37109375,
      "learning_rate": 0.0001893565336262773,
      "loss": 0.7068,
      "mean_token_accuracy": 0.8010134562849999,
      "num_tokens": 329386.0,
      "step": 140
    },
    {
      "entropy": 0.6653173223137856,
      "epoch": 0.1289490651192779,
      "grad_norm": 0.34375,
      "learning_rate": 0.00018746633197035527,
      "loss": 0.658,
      "mean_token_accuracy": 0.8119655668735504,
      "num_tokens": 349961.0,
      "step": 150
    },
    {
      "entropy": 0.7046581447124481,
      "epoch": 0.13754566946056307,
      "grad_norm": 0.3125,
      "learning_rate": 0.00018543279345021834,
      "loss": 0.7212,
      "mean_token_accuracy": 0.7986769005656242,
      "num_tokens": 373425.0,
      "step": 160
    },
    {
      "entropy": 0.7094632089138031,
      "epoch": 0.14614227380184827,
      "grad_norm": 0.333984375,
      "learning_rate": 0.00018325925055996076,
      "loss": 0.6972,
      "mean_token_accuracy": 0.8056794568896294,
      "num_tokens": 399788.0,
      "step": 170
    },
    {
      "entropy": 0.6955713860690593,
      "epoch": 0.15473887814313347,
      "grad_norm": 0.48046875,
      "learning_rate": 0.0001809492652280996,
      "loss": 0.6953,
      "mean_token_accuracy": 0.8080685377120972,
      "num_tokens": 421256.0,
      "step": 180
    },
    {
      "entropy": 0.672984279692173,
      "epoch": 0.16333548248441865,
      "grad_norm": 0.421875,
      "learning_rate": 0.00017850662298040678,
      "loss": 0.6758,
      "mean_token_accuracy": 0.8065061077475548,
      "num_tokens": 444196.0,
      "step": 190
    },
    {
      "entropy": 0.7029214389622211,
      "epoch": 0.17193208682570385,
      "grad_norm": 0.33984375,
      "learning_rate": 0.00017593532673631766,
      "loss": 0.7028,
      "mean_token_accuracy": 0.8078923970460892,
      "num_tokens": 468312.0,
      "step": 200
    },
    {
      "entropy": 0.6828642345964908,
      "epoch": 0.18052869116698905,
      "grad_norm": 0.431640625,
      "learning_rate": 0.00017323959024908209,
      "loss": 0.6942,
      "mean_token_accuracy": 0.8095744833350181,
      "num_tokens": 490571.0,
      "step": 210
    },
    {
      "entropy": 0.731064935028553,
      "epoch": 0.18912529550827423,
      "grad_norm": 0.37109375,
      "learning_rate": 0.00017042383120040834,
      "loss": 0.7514,
      "mean_token_accuracy": 0.8010679826140403,
      "num_tokens": 514374.0,
      "step": 220
    },
    {
      "entropy": 0.6761481508612632,
      "epoch": 0.19772189984955943,
      "grad_norm": 0.40234375,
      "learning_rate": 0.0001674926639609157,
      "loss": 0.6708,
      "mean_token_accuracy": 0.8103428333997726,
      "num_tokens": 536838.0,
      "step": 230
    },
    {
      "entropy": 0.7191244259476661,
      "epoch": 0.2063185041908446,
      "grad_norm": 0.3828125,
      "learning_rate": 0.0001644508920282601,
      "loss": 0.7343,
      "mean_token_accuracy": 0.7990544006228447,
      "num_tokens": 560798.0,
      "step": 240
    },
    {
      "entropy": 0.7184930846095086,
      "epoch": 0.2149151085321298,
      "grad_norm": 0.287109375,
      "learning_rate": 0.00016130350015532496,
      "loss": 0.7516,
      "mean_token_accuracy": 0.7994581028819084,
      "num_tokens": 586321.0,
      "step": 250
    },
    {
      "entropy": 0.7019176624715329,
      "epoch": 0.223511712873415,
      "grad_norm": 0.29296875,
      "learning_rate": 0.0001580556461813766,
      "loss": 0.7465,
      "mean_token_accuracy": 0.8056053712964057,
      "num_tokens": 612379.0,
      "step": 260
    },
    {
      "entropy": 0.7077236130833626,
      "epoch": 0.23210831721470018,
      "grad_norm": 0.408203125,
      "learning_rate": 0.00015471265257957202,
      "loss": 0.734,
      "mean_token_accuracy": 0.8064405977725982,
      "num_tokens": 637314.0,
      "step": 270
    },
    {
      "entropy": 0.7111990034580231,
      "epoch": 0.24070492155598538,
      "grad_norm": 0.318359375,
      "learning_rate": 0.00015127999773467002,
      "loss": 0.7056,
      "mean_token_accuracy": 0.806194306910038,
      "num_tokens": 660474.0,
      "step": 280
    },
    {
      "entropy": 0.7012526571750641,
      "epoch": 0.2493015258972706,
      "grad_norm": 0.3125,
      "learning_rate": 0.00014776330696523963,
      "loss": 0.7173,
      "mean_token_accuracy": 0.8059186190366745,
      "num_tokens": 684161.0,
      "step": 290
    },
    {
      "entropy": 0.7212809801101685,
      "epoch": 0.2578981302385558,
      "grad_norm": 0.337890625,
      "learning_rate": 0.00014416834330507856,
      "loss": 0.7445,
      "mean_token_accuracy": 0.7970411285758019,
      "num_tokens": 705287.0,
      "step": 300
    },
    {
      "entropy": 0.6817912630736828,
      "epoch": 0.26649473457984096,
      "grad_norm": 0.30078125,
      "learning_rate": 0.00014050099805894837,
      "loss": 0.6738,
      "mean_token_accuracy": 0.8141646087169647,
      "num_tokens": 728784.0,
      "step": 310
    },
    {
      "entropy": 0.7327330216765404,
      "epoch": 0.27509133892112614,
      "grad_norm": 0.400390625,
      "learning_rate": 0.00013676728114810367,
      "loss": 0.7564,
      "mean_token_accuracy": 0.7981535136699677,
      "num_tokens": 752938.0,
      "step": 320
    },
    {
      "entropy": 0.6538611635565758,
      "epoch": 0.28368794326241137,
      "grad_norm": 0.34765625,
      "learning_rate": 0.00013297331126143667,
      "loss": 0.6562,
      "mean_token_accuracy": 0.8156133487820625,
      "num_tokens": 776014.0,
      "step": 330
    },
    {
      "entropy": 0.6695164687931537,
      "epoch": 0.29228454760369654,
      "grad_norm": 0.400390625,
      "learning_rate": 0.00012912530582837682,
      "loss": 0.6904,
      "mean_token_accuracy": 0.8143788442015648,
      "num_tokens": 799890.0,
      "step": 340
    },
    {
      "entropy": 0.7247762531042099,
      "epoch": 0.3008811519449817,
      "grad_norm": 0.4296875,
      "learning_rate": 0.000125229570829978,
      "loss": 0.7753,
      "mean_token_accuracy": 0.8005941316485405,
      "num_tokens": 825088.0,
      "step": 350
    },
    {
      "entropy": 0.6589260399341583,
      "epoch": 0.30947775628626695,
      "grad_norm": 0.302734375,
      "learning_rate": 0.0001212924904648902,
      "loss": 0.6323,
      "mean_token_accuracy": 0.8167923167347908,
      "num_tokens": 849091.0,
      "step": 360
    },
    {
      "entropy": 0.6572450198233127,
      "epoch": 0.3180743606275521,
      "grad_norm": 0.427734375,
      "learning_rate": 0.00011732051668715081,
      "loss": 0.6631,
      "mean_token_accuracy": 0.8133938252925873,
      "num_tokens": 871233.0,
      "step": 370
    },
    {
      "entropy": 0.6870896026492119,
      "epoch": 0.3266709649688373,
      "grad_norm": 0.375,
      "learning_rate": 0.00011332015863294076,
      "loss": 0.7134,
      "mean_token_accuracy": 0.807905575633049,
      "num_tokens": 895100.0,
      "step": 380
    },
    {
      "entropy": 0.7120285458862782,
      "epoch": 0.3352675693101225,
      "grad_norm": 0.3203125,
      "learning_rate": 0.00010929797195363259,
      "loss": 0.7283,
      "mean_token_accuracy": 0.8010812431573868,
      "num_tokens": 917556.0,
      "step": 390
    },
    {
      "entropy": 0.6840016417205333,
      "epoch": 0.3438641736514077,
      "grad_norm": 0.3671875,
      "learning_rate": 0.00010526054807261067,
      "loss": 0.688,
      "mean_token_accuracy": 0.8081021830439568,
      "num_tokens": 939116.0,
      "step": 400
    }
  ],
  "logging_steps": 10,
  "max_steps": 800,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.4054474244501504e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}