{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9230769230769234,
  "eval_steps": 2,
  "global_step": 19,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 12.438282012939453,
      "learning_rate": 1e-05,
      "loss": 6.0391,
      "step": 1
    },
    {
      "epoch": 0.15384615384615385,
      "eval_loss": 5.750857830047607,
      "eval_runtime": 1.445,
      "eval_samples_per_second": 4.152,
      "eval_steps_per_second": 0.692,
      "step": 1
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 13.319328308105469,
      "learning_rate": 2e-05,
      "loss": 6.0695,
      "step": 2
    },
    {
      "epoch": 0.3076923076923077,
      "eval_loss": 5.732573986053467,
      "eval_runtime": 0.674,
      "eval_samples_per_second": 8.902,
      "eval_steps_per_second": 1.484,
      "step": 2
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 14.28262996673584,
      "learning_rate": 3e-05,
      "loss": 5.8912,
      "step": 3
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 12.073399543762207,
      "learning_rate": 4e-05,
      "loss": 5.7627,
      "step": 4
    },
    {
      "epoch": 0.6153846153846154,
      "eval_loss": 5.371483325958252,
      "eval_runtime": 0.6689,
      "eval_samples_per_second": 8.97,
      "eval_steps_per_second": 1.495,
      "step": 4
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 11.546090126037598,
      "learning_rate": 5e-05,
      "loss": 4.6946,
      "step": 5
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 8.326874732971191,
      "learning_rate": 6e-05,
      "loss": 4.5148,
      "step": 6
    },
    {
      "epoch": 0.9230769230769231,
      "eval_loss": 4.154338359832764,
      "eval_runtime": 0.6707,
      "eval_samples_per_second": 8.945,
      "eval_steps_per_second": 1.491,
      "step": 6
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 11.959038734436035,
      "learning_rate": 7e-05,
      "loss": 5.3493,
      "step": 7
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 6.133173942565918,
      "learning_rate": 8e-05,
      "loss": 3.3447,
      "step": 8
    },
    {
      "epoch": 1.2307692307692308,
      "eval_loss": 3.0085933208465576,
      "eval_runtime": 0.6669,
      "eval_samples_per_second": 8.996,
      "eval_steps_per_second": 1.499,
      "step": 8
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 8.133515357971191,
      "learning_rate": 9e-05,
      "loss": 3.0715,
      "step": 9
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 8.480439186096191,
      "learning_rate": 0.0001,
      "loss": 2.4508,
      "step": 10
    },
    {
      "epoch": 1.5384615384615383,
      "eval_loss": 1.6411386728286743,
      "eval_runtime": 0.6689,
      "eval_samples_per_second": 8.971,
      "eval_steps_per_second": 1.495,
      "step": 10
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 10.082879066467285,
      "learning_rate": 9.698463103929542e-05,
      "loss": 1.6102,
      "step": 11
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 14.025999069213867,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.9829,
      "step": 12
    },
    {
      "epoch": 1.8461538461538463,
      "eval_loss": 0.5054619908332825,
      "eval_runtime": 0.6703,
      "eval_samples_per_second": 8.951,
      "eval_steps_per_second": 1.492,
      "step": 12
    },
    {
      "epoch": 2.0,
      "grad_norm": 11.843927383422852,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.8831,
      "step": 13
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 5.380168437957764,
      "learning_rate": 5.868240888334653e-05,
      "loss": 0.2532,
      "step": 14
    },
    {
      "epoch": 2.1538461538461537,
      "eval_loss": 0.10065824538469315,
      "eval_runtime": 0.6735,
      "eval_samples_per_second": 8.908,
      "eval_steps_per_second": 1.485,
      "step": 14
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 2.8232905864715576,
      "learning_rate": 4.131759111665349e-05,
      "loss": 0.1064,
      "step": 15
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 1.3231079578399658,
      "learning_rate": 2.500000000000001e-05,
      "loss": 0.0464,
      "step": 16
    },
    {
      "epoch": 2.4615384615384617,
      "eval_loss": 0.024292172864079475,
      "eval_runtime": 0.6703,
      "eval_samples_per_second": 8.951,
      "eval_steps_per_second": 1.492,
      "step": 16
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 0.4727940261363983,
      "learning_rate": 1.1697777844051105e-05,
      "loss": 0.0198,
      "step": 17
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 0.27587053179740906,
      "learning_rate": 3.0153689607045845e-06,
      "loss": 0.0144,
      "step": 18
    },
    {
      "epoch": 2.769230769230769,
      "eval_loss": 0.013258029706776142,
      "eval_runtime": 0.6702,
      "eval_samples_per_second": 8.953,
      "eval_steps_per_second": 1.492,
      "step": 18
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 0.2778480648994446,
      "learning_rate": 0.0,
      "loss": 0.0125,
      "step": 19
    }
  ],
  "logging_steps": 1,
  "max_steps": 19,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.3724914745868288e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}