{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9770992366412212,
  "eval_steps": 500,
  "global_step": 390,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.153541599214077,
      "epoch": 0.07633587786259542,
      "grad_norm": 3.71875,
      "learning_rate": 3.461538461538462e-05,
      "loss": 0.5369,
      "mean_token_accuracy": 0.8995052814483643,
      "num_tokens": 915894.0,
      "step": 10
    },
    {
      "entropy": 0.9980269990861416,
      "epoch": 0.15267175572519084,
      "grad_norm": 1.9765625,
      "learning_rate": 7.307692307692307e-05,
      "loss": 0.0944,
      "mean_token_accuracy": 0.9686855398118496,
      "num_tokens": 1829473.0,
      "step": 20
    },
    {
      "entropy": 1.1567876577377318,
      "epoch": 0.22900763358778625,
      "grad_norm": 4.1875,
      "learning_rate": 9.998763518763365e-05,
      "loss": 0.1586,
      "mean_token_accuracy": 0.9656276367604733,
      "num_tokens": 2736915.0,
      "step": 30
    },
    {
      "entropy": 1.0751335889101028,
      "epoch": 0.3053435114503817,
      "grad_norm": 0.49609375,
      "learning_rate": 9.976804306460788e-05,
      "loss": 0.0682,
      "mean_token_accuracy": 0.9742538303136825,
      "num_tokens": 3644365.0,
      "step": 40
    },
    {
      "entropy": 1.0004250779747963,
      "epoch": 0.3816793893129771,
      "grad_norm": 0.40625,
      "learning_rate": 9.927552866136169e-05,
      "loss": 0.0579,
      "mean_token_accuracy": 0.9773690737783909,
      "num_tokens": 4554192.0,
      "step": 50
    },
    {
      "entropy": 1.0029304943978787,
      "epoch": 0.4580152671755725,
      "grad_norm": 0.578125,
      "learning_rate": 9.851369877189127e-05,
      "loss": 0.0612,
      "mean_token_accuracy": 0.9769754238426686,
      "num_tokens": 5454270.0,
      "step": 60
    },
    {
      "entropy": 1.0122574470937251,
      "epoch": 0.5343511450381679,
      "grad_norm": 0.44921875,
      "learning_rate": 9.74881324482068e-05,
      "loss": 0.0642,
      "mean_token_accuracy": 0.9744108475744724,
      "num_tokens": 6343987.0,
      "step": 70
    },
    {
      "entropy": 0.9858146071434021,
      "epoch": 0.6106870229007634,
      "grad_norm": 0.326171875,
      "learning_rate": 9.62063401436769e-05,
      "loss": 0.053,
      "mean_token_accuracy": 0.9782996632158756,
      "num_tokens": 7241839.0,
      "step": 80
    },
    {
      "entropy": 0.9321766041219235,
      "epoch": 0.6870229007633588,
      "grad_norm": 0.2890625,
      "learning_rate": 9.467770871228569e-05,
      "loss": 0.0502,
      "mean_token_accuracy": 0.9778276771306992,
      "num_tokens": 8149748.0,
      "step": 90
    },
    {
      "entropy": 0.9664321891963482,
      "epoch": 0.7633587786259542,
      "grad_norm": 0.322265625,
      "learning_rate": 9.29134326665851e-05,
      "loss": 0.0585,
      "mean_token_accuracy": 0.9770526088774204,
      "num_tokens": 9053166.0,
      "step": 100
    },
    {
      "entropy": 0.9390135675668716,
      "epoch": 0.8396946564885496,
      "grad_norm": 0.40625,
      "learning_rate": 9.092643219775594e-05,
      "loss": 0.0542,
      "mean_token_accuracy": 0.9784163929522037,
      "num_tokens": 9949074.0,
      "step": 110
    },
    {
      "entropy": 0.8867954194545746,
      "epoch": 0.916030534351145,
      "grad_norm": 0.28515625,
      "learning_rate": 8.87312585581356e-05,
      "loss": 0.0516,
      "mean_token_accuracy": 0.978719773888588,
      "num_tokens": 10850892.0,
      "step": 120
    },
    {
      "entropy": 0.8834987260401249,
      "epoch": 0.9923664122137404,
      "grad_norm": 0.40625,
      "learning_rate": 8.634398749911691e-05,
      "loss": 0.0539,
      "mean_token_accuracy": 0.9772126525640488,
      "num_tokens": 11751661.0,
      "step": 130
    },
    {
      "entropy": 0.8976177863776684,
      "epoch": 1.0687022900763359,
      "grad_norm": 0.36328125,
      "learning_rate": 8.378210154479702e-05,
      "loss": 0.0448,
      "mean_token_accuracy": 0.980753519386053,
      "num_tokens": 12646838.0,
      "step": 140
    },
    {
      "entropy": 0.8704435057938099,
      "epoch": 1.1450381679389312,
      "grad_norm": 0.333984375,
      "learning_rate": 8.106436196351265e-05,
      "loss": 0.0474,
      "mean_token_accuracy": 0.9792590379714966,
      "num_tokens": 13541728.0,
      "step": 150
    },
    {
      "entropy": 0.8593805082142353,
      "epoch": 1.2213740458015268,
      "grad_norm": 0.3125,
      "learning_rate": 7.821067137484279e-05,
      "loss": 0.0472,
      "mean_token_accuracy": 0.9803130567073822,
      "num_tokens": 14452617.0,
      "step": 160
    },
    {
      "entropy": 0.8790079154074192,
      "epoch": 1.297709923664122,
      "grad_norm": 0.3359375,
      "learning_rate": 7.524192799823925e-05,
      "loss": 0.0457,
      "mean_token_accuracy": 0.9807212583720684,
      "num_tokens": 15348770.0,
      "step": 170
    },
    {
      "entropy": 0.9099812150001526,
      "epoch": 1.3740458015267176,
      "grad_norm": 0.3125,
      "learning_rate": 7.217987261065473e-05,
      "loss": 0.0468,
      "mean_token_accuracy": 0.9811236143112183,
      "num_tokens": 16260924.0,
      "step": 180
    },
    {
      "entropy": 0.9149384334683418,
      "epoch": 1.450381679389313,
      "grad_norm": 0.28515625,
      "learning_rate": 6.904692933393251e-05,
      "loss": 0.0469,
      "mean_token_accuracy": 0.9806007631123066,
      "num_tokens": 17172435.0,
      "step": 190
    },
    {
      "entropy": 0.8875113040208816,
      "epoch": 1.5267175572519083,
      "grad_norm": 0.244140625,
      "learning_rate": 6.586604141790736e-05,
      "loss": 0.0413,
      "mean_token_accuracy": 0.9820852629840374,
      "num_tokens": 18072331.0,
      "step": 200
    },
    {
      "entropy": 0.8821073897182942,
      "epoch": 1.6030534351145038,
      "grad_norm": 0.34375,
      "learning_rate": 6.266050322181522e-05,
      "loss": 0.0463,
      "mean_token_accuracy": 0.9804246053099632,
      "num_tokens": 18986424.0,
      "step": 210
    },
    {
      "entropy": 0.8856651082634925,
      "epoch": 1.6793893129770994,
      "grad_norm": 0.30859375,
      "learning_rate": 5.945378962444953e-05,
      "loss": 0.0464,
      "mean_token_accuracy": 0.979961882531643,
      "num_tokens": 19881328.0,
      "step": 220
    },
    {
      "entropy": 0.8791401267051697,
      "epoch": 1.7557251908396947,
      "grad_norm": 0.25,
      "learning_rate": 5.6269384112332246e-05,
      "loss": 0.0447,
      "mean_token_accuracy": 0.9799509704113006,
      "num_tokens": 20777557.0,
      "step": 230
    },
    {
      "entropy": 0.8537083186209202,
      "epoch": 1.83206106870229,
      "grad_norm": 0.310546875,
      "learning_rate": 5.313060680484838e-05,
      "loss": 0.0434,
      "mean_token_accuracy": 0.9819154985249042,
      "num_tokens": 21679988.0,
      "step": 240
    },
    {
      "entropy": 0.8548756577074528,
      "epoch": 1.9083969465648853,
      "grad_norm": 0.2578125,
      "learning_rate": 5.006044367575511e-05,
      "loss": 0.0465,
      "mean_token_accuracy": 0.9799104005098342,
      "num_tokens": 22584307.0,
      "step": 250
    },
    {
      "entropy": 0.8446093052625656,
      "epoch": 1.984732824427481,
      "grad_norm": 0.271484375,
      "learning_rate": 4.708137822171435e-05,
      "loss": 0.045,
      "mean_token_accuracy": 0.981216999143362,
      "num_tokens": 23487511.0,
      "step": 260
    },
    {
      "entropy": 0.8299099013209343,
      "epoch": 2.0610687022900764,
      "grad_norm": 0.2216796875,
      "learning_rate": 4.4215226810578697e-05,
      "loss": 0.0388,
      "mean_token_accuracy": 0.9837547279894352,
      "num_tokens": 24391233.0,
      "step": 270
    },
    {
      "entropy": 0.7879593849182129,
      "epoch": 2.1374045801526718,
      "grad_norm": 0.2890625,
      "learning_rate": 4.1482978915211736e-05,
      "loss": 0.0401,
      "mean_token_accuracy": 0.982824681699276,
      "num_tokens": 25290459.0,
      "step": 280
    },
    {
      "entropy": 0.7747602254152298,
      "epoch": 2.213740458015267,
      "grad_norm": 0.279296875,
      "learning_rate": 3.890464340284692e-05,
      "loss": 0.0379,
      "mean_token_accuracy": 0.9832462161779404,
      "num_tokens": 26188999.0,
      "step": 290
    },
    {
      "entropy": 0.7617808029055595,
      "epoch": 2.2900763358778624,
      "grad_norm": 0.23828125,
      "learning_rate": 3.649910200564253e-05,
      "loss": 0.0386,
      "mean_token_accuracy": 0.9828567236661911,
      "num_tokens": 27102093.0,
      "step": 300
    },
    {
      "entropy": 0.7373107887804509,
      "epoch": 2.366412213740458,
      "grad_norm": 0.2392578125,
      "learning_rate": 3.42839710455007e-05,
      "loss": 0.0413,
      "mean_token_accuracy": 0.9825137570500374,
      "num_tokens": 28001749.0,
      "step": 310
    },
    {
      "entropy": 0.7243033863604069,
      "epoch": 2.4427480916030535,
      "grad_norm": 0.30078125,
      "learning_rate": 3.2275472425771183e-05,
      "loss": 0.038,
      "mean_token_accuracy": 0.9839053072035313,
      "num_tokens": 28899892.0,
      "step": 320
    },
    {
      "entropy": 0.7302201606333256,
      "epoch": 2.519083969465649,
      "grad_norm": 0.25390625,
      "learning_rate": 3.0488314834596443e-05,
      "loss": 0.0362,
      "mean_token_accuracy": 0.9841879785060883,
      "num_tokens": 29805646.0,
      "step": 330
    },
    {
      "entropy": 0.7226043172180653,
      "epoch": 2.595419847328244,
      "grad_norm": 0.33203125,
      "learning_rate": 2.8935586029872707e-05,
      "loss": 0.0396,
      "mean_token_accuracy": 0.9833728171885013,
      "num_tokens": 30704821.0,
      "step": 340
    },
    {
      "entropy": 0.7304494515061378,
      "epoch": 2.67175572519084,
      "grad_norm": 0.2275390625,
      "learning_rate": 2.7628656994648772e-05,
      "loss": 0.0385,
      "mean_token_accuracy": 0.9834493272006511,
      "num_tokens": 31612522.0,
      "step": 350
    },
    {
      "entropy": 0.7172894328832626,
      "epoch": 2.7480916030534353,
      "grad_norm": 0.228515625,
      "learning_rate": 2.6577098664853885e-05,
      "loss": 0.0368,
      "mean_token_accuracy": 0.9831421509385109,
      "num_tokens": 32509962.0,
      "step": 360
    },
    {
      "entropy": 0.7212371945381164,
      "epoch": 2.8244274809160306,
      "grad_norm": 0.326171875,
      "learning_rate": 2.5788611839176458e-05,
      "loss": 0.0391,
      "mean_token_accuracy": 0.9827444940805435,
      "num_tokens": 33422583.0,
      "step": 370
    },
    {
      "entropy": 0.7051329500973225,
      "epoch": 2.900763358778626,
      "grad_norm": 0.34375,
      "learning_rate": 2.5268970784379448e-05,
      "loss": 0.0398,
      "mean_token_accuracy": 0.9825506798923016,
      "num_tokens": 34314547.0,
      "step": 380
    },
    {
      "entropy": 0.7103805348277092,
      "epoch": 2.9770992366412212,
      "grad_norm": 0.3203125,
      "learning_rate": 2.5021980949043162e-05,
      "loss": 0.0401,
      "mean_token_accuracy": 0.9823778592050075,
      "num_tokens": 35226301.0,
      "step": 390
    }
  ],
  "logging_steps": 10,
  "max_steps": 393,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.502686600575416e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}