{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 1000,
  "global_step": 7591,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00039541320680110717,
      "grad_norm": 6.574801445007324,
      "learning_rate": 0.0,
      "loss": 4.3129,
      "step": 1
    },
    {
      "epoch": 0.39541320680110714,
      "grad_norm": 6.2015180587768555,
      "learning_rate": 9.871541501976284e-06,
      "loss": 3.298,
      "step": 1000
    },
    {
      "epoch": 0.39541320680110714,
      "eval_cosine_accuracy": 0.9471027255058289,
      "eval_loss": 0.5520303249359131,
      "eval_runtime": 36.3857,
      "eval_samples_per_second": 261.339,
      "eval_steps_per_second": 1.044,
      "step": 1000
    },
    {
      "epoch": 0.7908264136022143,
      "grad_norm": 8.382867813110352,
      "learning_rate": 1.9752964426877474e-05,
      "loss": 1.9673,
      "step": 2000
    },
    {
      "epoch": 0.7908264136022143,
      "eval_cosine_accuracy": 0.953938364982605,
      "eval_loss": 0.4995275139808655,
      "eval_runtime": 34.9654,
      "eval_samples_per_second": 271.955,
      "eval_steps_per_second": 1.087,
      "step": 2000
    },
    {
      "epoch": 1.1860924535756618,
      "grad_norm": 6.899052619934082,
      "learning_rate": 1.8930140597539545e-05,
      "loss": 1.5646,
      "step": 3000
    },
    {
      "epoch": 1.1860924535756618,
      "eval_cosine_accuracy": 0.9593017101287842,
      "eval_loss": 0.4840952455997467,
      "eval_runtime": 34.4403,
      "eval_samples_per_second": 276.101,
      "eval_steps_per_second": 1.103,
      "step": 3000
    },
    {
      "epoch": 1.581193204267088,
      "grad_norm": 6.6706132888793945,
      "learning_rate": 1.7831722319859405e-05,
      "loss": 1.5836,
      "step": 4000
    },
    {
      "epoch": 1.581193204267088,
      "eval_cosine_accuracy": 0.962666928768158,
      "eval_loss": 0.4643610715866089,
      "eval_runtime": 34.4996,
      "eval_samples_per_second": 275.627,
      "eval_steps_per_second": 1.101,
      "step": 4000
    },
    {
      "epoch": 1.9762939549585146,
      "grad_norm": 6.573458194732666,
      "learning_rate": 1.6733304042179265e-05,
      "loss": 1.4401,
      "step": 5000
    },
    {
      "epoch": 1.9762939549585146,
      "eval_cosine_accuracy": 0.9638237357139587,
      "eval_loss": 0.44956663250923157,
      "eval_runtime": 34.8836,
      "eval_samples_per_second": 272.593,
      "eval_steps_per_second": 1.089,
      "step": 5000
    },
    {
      "epoch": 2.3713947056499407,
      "grad_norm": 5.309902191162109,
      "learning_rate": 1.5634885764499125e-05,
      "loss": 1.2966,
      "step": 6000
    },
    {
      "epoch": 2.3713947056499407,
      "eval_cosine_accuracy": 0.9671889543533325,
      "eval_loss": 0.4553391933441162,
      "eval_runtime": 34.4231,
      "eval_samples_per_second": 276.239,
      "eval_steps_per_second": 1.104,
      "step": 6000
    },
    {
      "epoch": 2.7664954563413673,
      "grad_norm": 5.066314220428467,
      "learning_rate": 1.4538664323374343e-05,
      "loss": 1.2287,
      "step": 7000
    },
    {
      "epoch": 2.7664954563413673,
      "eval_cosine_accuracy": 0.9656115174293518,
      "eval_loss": 0.443572074174881,
      "eval_runtime": 34.6972,
      "eval_samples_per_second": 274.057,
      "eval_steps_per_second": 1.095,
      "step": 7000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 20232,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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
}