{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 1000,
  "global_step": 6704,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0002983293556085919,
      "grad_norm": 6.0359320640563965,
      "learning_rate": 0.0,
      "loss": 2.2793,
      "step": 1
    },
    {
      "epoch": 0.29832935560859186,
      "grad_norm": 3.87988543510437,
      "learning_rate": 5.960620525059666e-06,
      "loss": 2.265,
      "step": 1000
    },
    {
      "epoch": 0.29832935560859186,
      "eval_cosine_accuracy": 0.9628772735595703,
      "eval_loss": 0.8168219923973083,
      "eval_runtime": 23.679,
      "eval_samples_per_second": 401.579,
      "eval_steps_per_second": 1.605,
      "step": 1000
    },
    {
      "epoch": 0.5966587112171837,
      "grad_norm": 3.6651225090026855,
      "learning_rate": 1.1927207637231503e-05,
      "loss": 1.8687,
      "step": 2000
    },
    {
      "epoch": 0.5966587112171837,
      "eval_cosine_accuracy": 0.9732884764671326,
      "eval_loss": 0.7446231842041016,
      "eval_runtime": 23.5526,
      "eval_samples_per_second": 403.734,
      "eval_steps_per_second": 1.613,
      "step": 2000
    },
    {
      "epoch": 0.8949880668257757,
      "grad_norm": 3.408712863922119,
      "learning_rate": 1.7893794749403345e-05,
      "loss": 1.6653,
      "step": 3000
    },
    {
      "epoch": 0.8949880668257757,
      "eval_cosine_accuracy": 0.9795982837677002,
      "eval_loss": 0.6802141666412354,
      "eval_runtime": 23.5108,
      "eval_samples_per_second": 404.453,
      "eval_steps_per_second": 1.616,
      "step": 3000
    },
    {
      "epoch": 1.1933174224343674,
      "grad_norm": 3.534438133239746,
      "learning_rate": 1.903639618138425e-05,
      "loss": 1.4221,
      "step": 4000
    },
    {
      "epoch": 1.1933174224343674,
      "eval_cosine_accuracy": 0.9819118976593018,
      "eval_loss": 0.6609337329864502,
      "eval_runtime": 23.7912,
      "eval_samples_per_second": 399.686,
      "eval_steps_per_second": 1.597,
      "step": 4000
    },
    {
      "epoch": 1.4916467780429594,
      "grad_norm": 3.8688464164733887,
      "learning_rate": 1.754474940334129e-05,
      "loss": 1.4373,
      "step": 5000
    },
    {
      "epoch": 1.4916467780429594,
      "eval_cosine_accuracy": 0.9834893345832825,
      "eval_loss": 0.633538007736206,
      "eval_runtime": 23.5674,
      "eval_samples_per_second": 403.481,
      "eval_steps_per_second": 1.612,
      "step": 5000
    },
    {
      "epoch": 1.7899761336515514,
      "grad_norm": 3.5698401927948,
      "learning_rate": 1.6053102625298332e-05,
      "loss": 1.3735,
      "step": 6000
    },
    {
      "epoch": 1.7899761336515514,
      "eval_cosine_accuracy": 0.984015166759491,
      "eval_loss": 0.606587827205658,
      "eval_runtime": 24.689,
      "eval_samples_per_second": 385.151,
      "eval_steps_per_second": 1.539,
      "step": 6000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 16760,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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
}