{
  "best_global_step": 563,
  "best_metric": 1.3774118423461914,
  "best_model_checkpoint": "finetuned_models/FFT/checkpoint-563",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 563,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.5691047963500022,
      "epoch": 0.044444444444444446,
      "grad_norm": 61.75,
      "learning_rate": 2.840236686390533e-06,
      "loss": 2.82507080078125,
      "mean_token_accuracy": 0.5578909605741501,
      "num_tokens": 43581.0,
      "step": 25
    },
    {
      "entropy": 1.912336548268795,
      "epoch": 0.08888888888888889,
      "grad_norm": 41.25,
      "learning_rate": 5.7988165680473375e-06,
      "loss": 2.3563003540039062,
      "mean_token_accuracy": 0.5931275233626365,
      "num_tokens": 86889.0,
      "step": 50
    },
    {
      "entropy": 2.1532553315162657,
      "epoch": 0.13333333333333333,
      "grad_norm": 31.125,
      "learning_rate": 8.757396449704143e-06,
      "loss": 2.159627685546875,
      "mean_token_accuracy": 0.5959913417696953,
      "num_tokens": 130613.0,
      "step": 75
    },
    {
      "entropy": 1.9658770501613616,
      "epoch": 0.17777777777777778,
      "grad_norm": 30.0,
      "learning_rate": 1.1715976331360948e-05,
      "loss": 1.9642219543457031,
      "mean_token_accuracy": 0.6274332204461097,
      "num_tokens": 174274.0,
      "step": 100
    },
    {
      "entropy": 1.9181486976146698,
      "epoch": 0.2222222222222222,
      "grad_norm": 23.875,
      "learning_rate": 1.4674556213017752e-05,
      "loss": 1.9006834411621094,
      "mean_token_accuracy": 0.6362838646769524,
      "num_tokens": 217619.0,
      "step": 125
    },
    {
      "entropy": 1.8227408236265183,
      "epoch": 0.26666666666666666,
      "grad_norm": 29.5,
      "learning_rate": 1.7633136094674557e-05,
      "loss": 1.834572296142578,
      "mean_token_accuracy": 0.645097343325615,
      "num_tokens": 261407.0,
      "step": 150
    },
    {
      "entropy": 1.8350859326124191,
      "epoch": 0.3111111111111111,
      "grad_norm": 25.625,
      "learning_rate": 1.999946602771351e-05,
      "loss": 1.8147303771972656,
      "mean_token_accuracy": 0.6461810152232647,
      "num_tokens": 305402.0,
      "step": 175
    },
    {
      "entropy": 1.753722729086876,
      "epoch": 0.35555555555555557,
      "grad_norm": 31.5,
      "learning_rate": 1.9980782984658682e-05,
      "loss": 1.758114013671875,
      "mean_token_accuracy": 0.653534509241581,
      "num_tokens": 348998.0,
      "step": 200
    },
    {
      "entropy": 1.7685393619537353,
      "epoch": 0.4,
      "grad_norm": 23.75,
      "learning_rate": 1.9935458325191365e-05,
      "loss": 1.7483135986328124,
      "mean_token_accuracy": 0.6603708426654339,
      "num_tokens": 392837.0,
      "step": 225
    },
    {
      "entropy": 1.7378248673677446,
      "epoch": 0.4444444444444444,
      "grad_norm": 26.5,
      "learning_rate": 1.9863613034027224e-05,
      "loss": 1.7749085998535157,
      "mean_token_accuracy": 0.65810049533844,
      "num_tokens": 436281.0,
      "step": 250
    },
    {
      "entropy": 1.6766968676447869,
      "epoch": 0.4888888888888889,
      "grad_norm": 23.25,
      "learning_rate": 1.9765438887171327e-05,
      "loss": 1.668856658935547,
      "mean_token_accuracy": 0.6644712510704994,
      "num_tokens": 480451.0,
      "step": 275
    },
    {
      "entropy": 1.669287955760956,
      "epoch": 0.5333333333333333,
      "grad_norm": 20.625,
      "learning_rate": 1.9641197940012136e-05,
      "loss": 1.648075714111328,
      "mean_token_accuracy": 0.6689342276751995,
      "num_tokens": 524418.0,
      "step": 300
    },
    {
      "entropy": 1.587462457716465,
      "epoch": 0.5777777777777777,
      "grad_norm": 21.625,
      "learning_rate": 1.9491221827819348e-05,
      "loss": 1.5674880981445312,
      "mean_token_accuracy": 0.6790469540655613,
      "num_tokens": 568093.0,
      "step": 325
    },
    {
      "entropy": 1.5877643951773643,
      "epoch": 0.6222222222222222,
      "grad_norm": 23.5,
      "learning_rate": 1.9315910880512792e-05,
      "loss": 1.5833602905273438,
      "mean_token_accuracy": 0.67303916528821,
      "num_tokens": 611875.0,
      "step": 350
    },
    {
      "entropy": 1.5628123182058333,
      "epoch": 0.6666666666666666,
      "grad_norm": 19.875,
      "learning_rate": 1.911573305406528e-05,
      "loss": 1.5389964294433593,
      "mean_token_accuracy": 0.6916262510418892,
      "num_tokens": 655258.0,
      "step": 375
    },
    {
      "entropy": 1.5605447962880135,
      "epoch": 0.7111111111111111,
      "grad_norm": 25.25,
      "learning_rate": 1.8891222681391853e-05,
      "loss": 1.5674354553222656,
      "mean_token_accuracy": 0.6811535279452801,
      "num_tokens": 698915.0,
      "step": 400
    },
    {
      "entropy": 1.5302811539173127,
      "epoch": 0.7555555555555555,
      "grad_norm": 21.0,
      "learning_rate": 1.8642979046059595e-05,
      "loss": 1.5149029541015624,
      "mean_token_accuracy": 0.6928434443473815,
      "num_tokens": 742608.0,
      "step": 425
    },
    {
      "entropy": 1.4984435787796975,
      "epoch": 0.8,
      "grad_norm": 20.5,
      "learning_rate": 1.8371664782625287e-05,
      "loss": 1.4783544921875,
      "mean_token_accuracy": 0.6922086542844772,
      "num_tokens": 785924.0,
      "step": 450
    },
    {
      "entropy": 1.4762829530239105,
      "epoch": 0.8444444444444444,
      "grad_norm": 22.875,
      "learning_rate": 1.8078004107870797e-05,
      "loss": 1.4516114807128906,
      "mean_token_accuracy": 0.702991289794445,
      "num_tokens": 829069.0,
      "step": 475
    },
    {
      "entropy": 1.5141191801428795,
      "epoch": 0.8888888888888888,
      "grad_norm": 20.75,
      "learning_rate": 1.7762780887657576e-05,
      "loss": 1.5113209533691405,
      "mean_token_accuracy": 0.6913984973728656,
      "num_tokens": 872824.0,
      "step": 500
    },
    {
      "entropy": 1.4147853454947472,
      "epoch": 0.9333333333333333,
      "grad_norm": 25.75,
      "learning_rate": 1.74268365445604e-05,
      "loss": 1.394475555419922,
      "mean_token_accuracy": 0.712423055768013,
      "num_tokens": 916212.0,
      "step": 525
    },
    {
      "entropy": 1.4362201416492462,
      "epoch": 0.9777777777777777,
      "grad_norm": 21.25,
      "learning_rate": 1.7071067811865477e-05,
      "loss": 1.4004106140136718,
      "mean_token_accuracy": 0.7106147593259812,
      "num_tokens": 959695.0,
      "step": 550
    },
    {
      "epoch": 1.0,
      "eval_entropy": 1.3886345291137696,
      "eval_loss": 1.3774118423461914,
      "eval_mean_token_accuracy": 0.7105021185874939,
      "eval_num_tokens": 981504.0,
      "eval_runtime": 28.7124,
      "eval_samples_per_second": 34.828,
      "eval_steps_per_second": 4.354,
      "step": 563
    }
  ],
  "logging_steps": 25,
  "max_steps": 1689,
  "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": 2522133763928064.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}