{
  "best_global_step": null,
  "best_metric": Infinity,
  "best_model_checkpoint": null,
  "epoch": 0.6097560975609756,
  "eval_steps": 200,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.0894674882292747,
      "epoch": 0.001524390243902439,
      "grad_norm": 0.2451171875,
      "learning_rate": 0.0,
      "loss": 1.1132,
      "mean_token_accuracy": 0.737787663936615,
      "num_tokens": 31096.0,
      "step": 1
    },
    {
      "entropy": 1.1255738629322303,
      "epoch": 0.03048780487804878,
      "grad_norm": 0.259765625,
      "learning_rate": 8.260869565217392e-07,
      "loss": 1.0807,
      "mean_token_accuracy": 0.7315414132256257,
      "num_tokens": 610285.0,
      "step": 20
    },
    {
      "entropy": 1.116651164740324,
      "epoch": 0.06097560975609756,
      "grad_norm": 0.2060546875,
      "learning_rate": 1.6956521739130435e-06,
      "loss": 1.0829,
      "mean_token_accuracy": 0.7415120448917151,
      "num_tokens": 1213808.0,
      "step": 40
    },
    {
      "entropy": 1.136062316596508,
      "epoch": 0.09146341463414634,
      "grad_norm": 0.255859375,
      "learning_rate": 2.5652173913043484e-06,
      "loss": 1.0719,
      "mean_token_accuracy": 0.7423151686787606,
      "num_tokens": 1820350.0,
      "step": 60
    },
    {
      "entropy": 1.131455584987998,
      "epoch": 0.12195121951219512,
      "grad_norm": 0.2490234375,
      "learning_rate": 3.4347826086956526e-06,
      "loss": 1.0751,
      "mean_token_accuracy": 0.7383749138563871,
      "num_tokens": 2425417.0,
      "step": 80
    },
    {
      "entropy": 1.1437791034579277,
      "epoch": 0.1524390243902439,
      "grad_norm": 0.2451171875,
      "learning_rate": 4.304347826086957e-06,
      "loss": 1.0587,
      "mean_token_accuracy": 0.7454101398587227,
      "num_tokens": 3028789.0,
      "step": 100
    },
    {
      "entropy": 1.1516818165779115,
      "epoch": 0.18292682926829268,
      "grad_norm": 0.267578125,
      "learning_rate": 5.173913043478262e-06,
      "loss": 1.0153,
      "mean_token_accuracy": 0.7389566078782082,
      "num_tokens": 3633111.0,
      "step": 120
    },
    {
      "entropy": 1.1731348514556885,
      "epoch": 0.21341463414634146,
      "grad_norm": 0.2119140625,
      "learning_rate": 6.043478260869565e-06,
      "loss": 0.9987,
      "mean_token_accuracy": 0.7419542625546456,
      "num_tokens": 4238255.0,
      "step": 140
    },
    {
      "entropy": 1.172143243253231,
      "epoch": 0.24390243902439024,
      "grad_norm": 0.224609375,
      "learning_rate": 6.91304347826087e-06,
      "loss": 0.9389,
      "mean_token_accuracy": 0.7644132591784001,
      "num_tokens": 4832556.0,
      "step": 160
    },
    {
      "entropy": 1.2004285369068384,
      "epoch": 0.27439024390243905,
      "grad_norm": 0.17578125,
      "learning_rate": 7.782608695652174e-06,
      "loss": 0.9192,
      "mean_token_accuracy": 0.7614704228937625,
      "num_tokens": 5432124.0,
      "step": 180
    },
    {
      "entropy": 1.2147842556238175,
      "epoch": 0.3048780487804878,
      "grad_norm": 0.21484375,
      "learning_rate": 8.65217391304348e-06,
      "loss": 0.8948,
      "mean_token_accuracy": 0.7692887011915446,
      "num_tokens": 6037582.0,
      "step": 200
    },
    {
      "epoch": 0.3048780487804878,
      "eval_entropy": 1.2062757316433237,
      "eval_loss": NaN,
      "eval_mean_token_accuracy": 0.7701100997653296,
      "eval_num_tokens": 6037582.0,
      "eval_runtime": 226.3543,
      "eval_samples_per_second": 2.483,
      "eval_steps_per_second": 1.241,
      "step": 200
    },
    {
      "entropy": 1.209118977189064,
      "epoch": 0.3353658536585366,
      "grad_norm": 0.166015625,
      "learning_rate": 9.521739130434784e-06,
      "loss": 0.8307,
      "mean_token_accuracy": 0.7851326018571854,
      "num_tokens": 6638022.0,
      "step": 220
    },
    {
      "entropy": 1.2081195436418057,
      "epoch": 0.36585365853658536,
      "grad_norm": 0.154296875,
      "learning_rate": 1.0391304347826088e-05,
      "loss": 0.8054,
      "mean_token_accuracy": 0.775788264721632,
      "num_tokens": 7242603.0,
      "step": 240
    },
    {
      "entropy": 1.218064185231924,
      "epoch": 0.39634146341463417,
      "grad_norm": 0.1572265625,
      "learning_rate": 1.1260869565217392e-05,
      "loss": 0.8023,
      "mean_token_accuracy": 0.7786701653152704,
      "num_tokens": 7854134.0,
      "step": 260
    },
    {
      "entropy": 1.222281923890114,
      "epoch": 0.4268292682926829,
      "grad_norm": 0.2080078125,
      "learning_rate": 1.2130434782608698e-05,
      "loss": 0.7419,
      "mean_token_accuracy": 0.7926970385015011,
      "num_tokens": 8456300.0,
      "step": 280
    },
    {
      "entropy": 1.1952788300812245,
      "epoch": 0.4573170731707317,
      "grad_norm": 0.1982421875,
      "learning_rate": 1.3000000000000001e-05,
      "loss": 0.7198,
      "mean_token_accuracy": 0.800191256031394,
      "num_tokens": 9058721.0,
      "step": 300
    },
    {
      "entropy": 1.2105496317148208,
      "epoch": 0.4878048780487805,
      "grad_norm": 0.228515625,
      "learning_rate": 1.3869565217391305e-05,
      "loss": 0.6962,
      "mean_token_accuracy": 0.8068932801485061,
      "num_tokens": 9662758.0,
      "step": 320
    },
    {
      "entropy": 1.2259060773998498,
      "epoch": 0.5182926829268293,
      "grad_norm": 0.189453125,
      "learning_rate": 1.473913043478261e-05,
      "loss": 0.7312,
      "mean_token_accuracy": 0.8055971477180719,
      "num_tokens": 10271738.0,
      "step": 340
    },
    {
      "entropy": 1.2056082211434842,
      "epoch": 0.5487804878048781,
      "grad_norm": 0.1826171875,
      "learning_rate": 1.5608695652173914e-05,
      "loss": 0.6527,
      "mean_token_accuracy": 0.7986662428826093,
      "num_tokens": 10879740.0,
      "step": 360
    },
    {
      "entropy": 1.225981270521879,
      "epoch": 0.5792682926829268,
      "grad_norm": 0.259765625,
      "learning_rate": 1.6478260869565218e-05,
      "loss": 0.682,
      "mean_token_accuracy": 0.7959255930036306,
      "num_tokens": 11483336.0,
      "step": 380
    },
    {
      "entropy": 1.1928005013614893,
      "epoch": 0.6097560975609756,
      "grad_norm": 0.201171875,
      "learning_rate": 1.7347826086956522e-05,
      "loss": 0.6594,
      "mean_token_accuracy": 0.8185551591217518,
      "num_tokens": 12085058.0,
      "step": 400
    },
    {
      "epoch": 0.6097560975609756,
      "eval_entropy": 1.1817513504910724,
      "eval_loss": NaN,
      "eval_mean_token_accuracy": 0.806651482590577,
      "eval_num_tokens": 12085058.0,
      "eval_runtime": 229.1412,
      "eval_samples_per_second": 2.453,
      "eval_steps_per_second": 1.226,
      "step": 400
    }
  ],
  "logging_steps": 20,
  "max_steps": 4592,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 5.5195257431906304e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}