{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 1000,
  "global_step": 3120,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0003205128205128205,
      "grad_norm": 5.640012741088867,
      "learning_rate": 0.0,
      "loss": 2.1128,
      "step": 1
    },
    {
      "epoch": 0.32051282051282054,
      "grad_norm": 3.1269683837890625,
      "learning_rate": 2.9775641025641028e-05,
      "loss": 1.6035,
      "step": 1000
    },
    {
      "epoch": 0.32051282051282054,
      "eval_cosine_accuracy": 0.9652960300445557,
      "eval_loss": 0.7382573485374451,
      "eval_runtime": 29.5289,
      "eval_samples_per_second": 322.024,
      "eval_steps_per_second": 1.287,
      "step": 1000
    },
    {
      "epoch": 0.6410256410256411,
      "grad_norm": 3.351823091506958,
      "learning_rate": 2.6214387464387465e-05,
      "loss": 1.0575,
      "step": 2000
    },
    {
      "epoch": 0.6410256410256411,
      "eval_cosine_accuracy": 0.9737091064453125,
      "eval_loss": 0.6786530017852783,
      "eval_runtime": 28.6588,
      "eval_samples_per_second": 331.801,
      "eval_steps_per_second": 1.326,
      "step": 2000
    },
    {
      "epoch": 0.9615384615384616,
      "grad_norm": 2.9926202297210693,
      "learning_rate": 2.2653133903133903e-05,
      "loss": 1.8049,
      "step": 3000
    },
    {
      "epoch": 0.9615384615384616,
      "eval_cosine_accuracy": 0.9739194512367249,
      "eval_loss": 0.659064531326294,
      "eval_runtime": 28.2842,
      "eval_samples_per_second": 336.195,
      "eval_steps_per_second": 1.344,
      "step": 3000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 9360,
  "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": 0.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}