{
  "best_metric": 0.8664895296096802,
  "best_model_checkpoint": "miner_id_24/checkpoint-25",
  "epoch": 3.0476190476190474,
  "eval_steps": 25,
  "global_step": 40,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0761904761904762,
      "grad_norm": 13.793068885803223,
      "learning_rate": 5e-05,
      "loss": 7.5161,
      "step": 1
    },
    {
      "epoch": 0.0761904761904762,
      "eval_loss": 7.238284111022949,
      "eval_runtime": 1.7184,
      "eval_samples_per_second": 51.793,
      "eval_steps_per_second": 6.983,
      "step": 1
    },
    {
      "epoch": 0.1523809523809524,
      "grad_norm": 14.815467834472656,
      "learning_rate": 0.0001,
      "loss": 7.9133,
      "step": 2
    },
    {
      "epoch": 0.22857142857142856,
      "grad_norm": 13.63614559173584,
      "learning_rate": 9.98292246503335e-05,
      "loss": 7.1754,
      "step": 3
    },
    {
      "epoch": 0.3047619047619048,
      "grad_norm": 11.953235626220703,
      "learning_rate": 9.931806517013612e-05,
      "loss": 5.6348,
      "step": 4
    },
    {
      "epoch": 0.38095238095238093,
      "grad_norm": 9.636924743652344,
      "learning_rate": 9.847001329696653e-05,
      "loss": 4.3034,
      "step": 5
    },
    {
      "epoch": 0.45714285714285713,
      "grad_norm": 9.072868347167969,
      "learning_rate": 9.729086208503174e-05,
      "loss": 3.2784,
      "step": 6
    },
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 7.585874080657959,
      "learning_rate": 9.578866633275288e-05,
      "loss": 2.3827,
      "step": 7
    },
    {
      "epoch": 0.6095238095238096,
      "grad_norm": 7.189624309539795,
      "learning_rate": 9.397368756032445e-05,
      "loss": 1.9073,
      "step": 8
    },
    {
      "epoch": 0.6857142857142857,
      "grad_norm": 4.020777225494385,
      "learning_rate": 9.185832391312644e-05,
      "loss": 1.2588,
      "step": 9
    },
    {
      "epoch": 0.7619047619047619,
      "grad_norm": 2.3023617267608643,
      "learning_rate": 8.945702546981969e-05,
      "loss": 1.0257,
      "step": 10
    },
    {
      "epoch": 0.8380952380952381,
      "grad_norm": 3.1915535926818848,
      "learning_rate": 8.678619553365659e-05,
      "loss": 0.9981,
      "step": 11
    },
    {
      "epoch": 0.9142857142857143,
      "grad_norm": 1.108399748802185,
      "learning_rate": 8.386407858128706e-05,
      "loss": 0.9343,
      "step": 12
    },
    {
      "epoch": 0.9904761904761905,
      "grad_norm": 1.5387076139450073,
      "learning_rate": 8.07106356344834e-05,
      "loss": 0.9265,
      "step": 13
    },
    {
      "epoch": 1.0666666666666667,
      "grad_norm": 5.272054195404053,
      "learning_rate": 7.734740790612136e-05,
      "loss": 1.6916,
      "step": 14
    },
    {
      "epoch": 1.1428571428571428,
      "grad_norm": 3.3526999950408936,
      "learning_rate": 7.379736965185368e-05,
      "loss": 0.9047,
      "step": 15
    },
    {
      "epoch": 1.2190476190476192,
      "grad_norm": 0.9935436248779297,
      "learning_rate": 7.008477123264848e-05,
      "loss": 0.9146,
      "step": 16
    },
    {
      "epoch": 1.2952380952380953,
      "grad_norm": 1.9652646780014038,
      "learning_rate": 6.623497346023418e-05,
      "loss": 0.886,
      "step": 17
    },
    {
      "epoch": 1.3714285714285714,
      "grad_norm": 1.2924121618270874,
      "learning_rate": 6.227427435703997e-05,
      "loss": 0.9051,
      "step": 18
    },
    {
      "epoch": 1.4476190476190476,
      "grad_norm": 1.112022042274475,
      "learning_rate": 5.8229729514036705e-05,
      "loss": 0.9189,
      "step": 19
    },
    {
      "epoch": 1.5238095238095237,
      "grad_norm": 0.6629959344863892,
      "learning_rate": 5.4128967273616625e-05,
      "loss": 0.8658,
      "step": 20
    },
    {
      "epoch": 1.6,
      "grad_norm": 1.3960295915603638,
      "learning_rate": 5e-05,
      "loss": 0.9024,
      "step": 21
    },
    {
      "epoch": 1.6761904761904762,
      "grad_norm": 1.09142005443573,
      "learning_rate": 4.5871032726383386e-05,
      "loss": 0.8813,
      "step": 22
    },
    {
      "epoch": 1.7523809523809524,
      "grad_norm": 0.8525294065475464,
      "learning_rate": 4.17702704859633e-05,
      "loss": 0.8702,
      "step": 23
    },
    {
      "epoch": 1.8285714285714287,
      "grad_norm": 1.6207659244537354,
      "learning_rate": 3.772572564296005e-05,
      "loss": 0.9112,
      "step": 24
    },
    {
      "epoch": 1.9047619047619047,
      "grad_norm": 0.6471120715141296,
      "learning_rate": 3.3765026539765834e-05,
      "loss": 0.8969,
      "step": 25
    },
    {
      "epoch": 1.9047619047619047,
      "eval_loss": 0.8664895296096802,
      "eval_runtime": 1.7228,
      "eval_samples_per_second": 51.662,
      "eval_steps_per_second": 6.966,
      "step": 25
    },
    {
      "epoch": 1.980952380952381,
      "grad_norm": 0.5163573622703552,
      "learning_rate": 2.991522876735154e-05,
      "loss": 0.8898,
      "step": 26
    },
    {
      "epoch": 2.057142857142857,
      "grad_norm": 1.372274398803711,
      "learning_rate": 2.6202630348146324e-05,
      "loss": 1.63,
      "step": 27
    },
    {
      "epoch": 2.1333333333333333,
      "grad_norm": 0.9707751274108887,
      "learning_rate": 2.2652592093878666e-05,
      "loss": 0.8799,
      "step": 28
    },
    {
      "epoch": 2.2095238095238097,
      "grad_norm": 0.9621633291244507,
      "learning_rate": 1.928936436551661e-05,
      "loss": 0.8554,
      "step": 29
    },
    {
      "epoch": 2.2857142857142856,
      "grad_norm": 0.5492159128189087,
      "learning_rate": 1.6135921418712956e-05,
      "loss": 0.8722,
      "step": 30
    },
    {
      "epoch": 2.361904761904762,
      "grad_norm": 0.5442349314689636,
      "learning_rate": 1.3213804466343421e-05,
      "loss": 0.8885,
      "step": 31
    },
    {
      "epoch": 2.4380952380952383,
      "grad_norm": 0.5701318383216858,
      "learning_rate": 1.0542974530180327e-05,
      "loss": 0.8383,
      "step": 32
    },
    {
      "epoch": 2.5142857142857142,
      "grad_norm": 0.5509917736053467,
      "learning_rate": 8.141676086873572e-06,
      "loss": 0.8445,
      "step": 33
    },
    {
      "epoch": 2.5904761904761906,
      "grad_norm": 0.4586718678474426,
      "learning_rate": 6.026312439675552e-06,
      "loss": 0.8689,
      "step": 34
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.5567528009414673,
      "learning_rate": 4.2113336672471245e-06,
      "loss": 0.8582,
      "step": 35
    },
    {
      "epoch": 2.742857142857143,
      "grad_norm": 1.0045620203018188,
      "learning_rate": 2.7091379149682685e-06,
      "loss": 0.8702,
      "step": 36
    },
    {
      "epoch": 2.819047619047619,
      "grad_norm": 0.47573980689048767,
      "learning_rate": 1.5299867030334814e-06,
      "loss": 0.8581,
      "step": 37
    },
    {
      "epoch": 2.895238095238095,
      "grad_norm": 0.525488555431366,
      "learning_rate": 6.819348298638839e-07,
      "loss": 0.8556,
      "step": 38
    },
    {
      "epoch": 2.9714285714285715,
      "grad_norm": 0.7117540836334229,
      "learning_rate": 1.7077534966650766e-07,
      "loss": 0.857,
      "step": 39
    },
    {
      "epoch": 3.0476190476190474,
      "grad_norm": 2.8340647220611572,
      "learning_rate": 0.0,
      "loss": 1.6132,
      "step": 40
    }
  ],
  "logging_steps": 1,
  "max_steps": 40,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 25,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 1,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.2528685890404352e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}