{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.502749410840534,
  "eval_steps": 100,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.4948265369981528,
      "epoch": 0.12568735271013354,
      "grad_norm": 0.0615234375,
      "learning_rate": 0.00015000000000000001,
      "loss": 1.7029500961303712,
      "mean_token_accuracy": 0.6537273893132806,
      "num_tokens": 156024.0,
      "step": 10
    },
    {
      "entropy": 1.1234326036646962,
      "epoch": 0.2513747054202671,
      "grad_norm": 0.037353515625,
      "learning_rate": 0.00019953520716943371,
      "loss": 1.1771140098571777,
      "mean_token_accuracy": 0.7510503690689803,
      "num_tokens": 298620.0,
      "step": 20
    },
    {
      "entropy": 1.0369394151493907,
      "epoch": 0.3770620581304006,
      "grad_norm": 0.0267333984375,
      "learning_rate": 0.0001972690659618564,
      "loss": 1.1110390663146972,
      "mean_token_accuracy": 0.7662029687315226,
      "num_tokens": 449966.0,
      "step": 30
    },
    {
      "entropy": 1.0444004433229566,
      "epoch": 0.5027494108405341,
      "grad_norm": 0.02880859375,
      "learning_rate": 0.0001931591088051279,
      "loss": 1.1269343376159668,
      "mean_token_accuracy": 0.7651370905339718,
      "num_tokens": 608754.0,
      "step": 40
    },
    {
      "entropy": 1.051739121414721,
      "epoch": 0.6284367635506677,
      "grad_norm": 0.0255126953125,
      "learning_rate": 0.00018728324335139814,
      "loss": 1.088887882232666,
      "mean_token_accuracy": 0.7666051037609577,
      "num_tokens": 764583.0,
      "step": 50
    },
    {
      "entropy": 0.9839291835203767,
      "epoch": 0.7541241162608012,
      "grad_norm": 0.029052734375,
      "learning_rate": 0.0001797528515115709,
      "loss": 1.020584487915039,
      "mean_token_accuracy": 0.7803368698805571,
      "num_tokens": 912716.0,
      "step": 60
    },
    {
      "entropy": 1.0362637933343648,
      "epoch": 0.8798114689709348,
      "grad_norm": 0.0244140625,
      "learning_rate": 0.00017071067811865476,
      "loss": 1.0753373146057128,
      "mean_token_accuracy": 0.7713750531896949,
      "num_tokens": 1064494.0,
      "step": 70
    },
    {
      "entropy": 0.9550455894345552,
      "epoch": 1.0,
      "grad_norm": 0.02880859375,
      "learning_rate": 0.0001603281250808719,
      "loss": 0.9675676345825195,
      "mean_token_accuracy": 0.7857394160008898,
      "num_tokens": 1201271.0,
      "step": 80
    },
    {
      "entropy": 0.9374621393159032,
      "epoch": 1.1256873527101336,
      "grad_norm": 0.0244140625,
      "learning_rate": 0.00014880200231609983,
      "loss": 0.9870312690734864,
      "mean_token_accuracy": 0.7901170210912823,
      "num_tokens": 1349133.0,
      "step": 90
    },
    {
      "entropy": 0.949882148578763,
      "epoch": 1.251374705420267,
      "grad_norm": 0.03173828125,
      "learning_rate": 0.00013635079705638298,
      "loss": 0.9436046600341796,
      "mean_token_accuracy": 0.788494935259223,
      "num_tokens": 1500927.0,
      "step": 100
    },
    {
      "epoch": 1.251374705420267,
      "eval_entropy": 0.9404270783276625,
      "eval_loss": 0.9446325302124023,
      "eval_mean_token_accuracy": 0.7829881757497787,
      "eval_num_tokens": 1500927.0,
      "eval_runtime": 86.1079,
      "eval_samples_per_second": 1.649,
      "eval_steps_per_second": 1.649,
      "step": 100
    },
    {
      "entropy": 0.9173299714922905,
      "epoch": 1.3770620581304005,
      "grad_norm": 0.0233154296875,
      "learning_rate": 0.0001232105322409468,
      "loss": 0.9481925964355469,
      "mean_token_accuracy": 0.7930883213877677,
      "num_tokens": 1646678.0,
      "step": 110
    },
    {
      "entropy": 0.9073959412053227,
      "epoch": 1.5027494108405341,
      "grad_norm": 0.03076171875,
      "learning_rate": 0.00010963029250531418,
      "loss": 0.9351880073547363,
      "mean_token_accuracy": 0.7968914289027452,
      "num_tokens": 1799797.0,
      "step": 120
    },
    {
      "entropy": 0.9462396390736103,
      "epoch": 1.6284367635506678,
      "grad_norm": 0.029296875,
      "learning_rate": 9.586750257511867e-05,
      "loss": 0.9954720497131347,
      "mean_token_accuracy": 0.789978607185185,
      "num_tokens": 1958692.0,
      "step": 130
    },
    {
      "entropy": 0.9035223769024014,
      "epoch": 1.7541241162608012,
      "grad_norm": 0.02880859375,
      "learning_rate": 8.218304756658072e-05,
      "loss": 0.9473580360412598,
      "mean_token_accuracy": 0.7965094247832895,
      "num_tokens": 2098856.0,
      "step": 140
    },
    {
      "entropy": 0.9002945913001895,
      "epoch": 1.8798114689709347,
      "grad_norm": 0.035888671875,
      "learning_rate": 6.883632769240589e-05,
      "loss": 0.9326913833618165,
      "mean_token_accuracy": 0.800568002089858,
      "num_tokens": 2253582.0,
      "step": 150
    },
    {
      "entropy": 0.8597502765897053,
      "epoch": 2.0,
      "grad_norm": 0.031494140625,
      "learning_rate": 5.608034111526298e-05,
      "loss": 0.8665840148925781,
      "mean_token_accuracy": 0.8103327634287816,
      "num_tokens": 2402542.0,
      "step": 160
    },
    {
      "entropy": 0.8727964337915182,
      "epoch": 2.1256873527101336,
      "grad_norm": 0.03173828125,
      "learning_rate": 4.415688815743858e-05,
      "loss": 0.9112902641296386,
      "mean_token_accuracy": 0.8074018105864524,
      "num_tokens": 2554831.0,
      "step": 170
    },
    {
      "entropy": 0.8327508143149316,
      "epoch": 2.2513747054202673,
      "grad_norm": 0.03955078125,
      "learning_rate": 3.329198777485869e-05,
      "loss": 0.8397786140441894,
      "mean_token_accuracy": 0.816638495773077,
      "num_tokens": 2700898.0,
      "step": 180
    },
    {
      "entropy": 0.8549322345294058,
      "epoch": 2.3770620581304005,
      "grad_norm": 0.041015625,
      "learning_rate": 2.3691593180019366e-05,
      "loss": 0.8526265144348144,
      "mean_token_accuracy": 0.8113520180806517,
      "num_tokens": 2854824.0,
      "step": 190
    },
    {
      "entropy": 0.8102049398235976,
      "epoch": 2.502749410840534,
      "grad_norm": 0.06787109375,
      "learning_rate": 1.553768782775351e-05,
      "loss": 0.8202457427978516,
      "mean_token_accuracy": 0.8225593730807305,
      "num_tokens": 3003572.0,
      "step": 200
    },
    {
      "epoch": 2.502749410840534,
      "eval_entropy": 0.8503286918284187,
      "eval_loss": 0.8580905795097351,
      "eval_mean_token_accuracy": 0.8025841964802272,
      "eval_num_tokens": 3003572.0,
      "eval_runtime": 85.9551,
      "eval_samples_per_second": 1.652,
      "eval_steps_per_second": 1.652,
      "step": 200
    }
  ],
  "logging_steps": 10,
  "max_steps": 240,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 1.2965602468164403e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}