{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 1000,
  "global_step": 7056,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0002835270768358378,
      "grad_norm": 5.961081027984619,
      "learning_rate": 0.0,
      "loss": 2.7767,
      "step": 1
    },
    {
      "epoch": 0.28352707683583783,
      "grad_norm": 3.903580665588379,
      "learning_rate": 2.8300283286118982e-05,
      "loss": 2.1777,
      "step": 1000
    },
    {
      "epoch": 0.28352707683583783,
      "eval_cosine_accuracy": 0.9602788686752319,
      "eval_loss": 1.0198065042495728,
      "eval_runtime": 28.5803,
      "eval_samples_per_second": 331.207,
      "eval_steps_per_second": 1.295,
      "step": 1000
    },
    {
      "epoch": 0.5670541536716757,
      "grad_norm": 3.8064451217651367,
      "learning_rate": 2.7041587901701324e-05,
      "loss": 1.5725,
      "step": 2000
    },
    {
      "epoch": 0.5670541536716757,
      "eval_cosine_accuracy": 0.967145562171936,
      "eval_loss": 0.9300715327262878,
      "eval_runtime": 27.2211,
      "eval_samples_per_second": 347.745,
      "eval_steps_per_second": 1.359,
      "step": 2000
    },
    {
      "epoch": 0.8505812305075134,
      "grad_norm": 3.4235427379608154,
      "learning_rate": 2.3890989287964716e-05,
      "loss": 1.472,
      "step": 3000
    },
    {
      "epoch": 0.8505812305075134,
      "eval_cosine_accuracy": 0.9721106886863708,
      "eval_loss": 0.847688615322113,
      "eval_runtime": 33.136,
      "eval_samples_per_second": 285.671,
      "eval_steps_per_second": 1.117,
      "step": 3000
    },
    {
      "epoch": 1.134032303768773,
      "grad_norm": 3.523313283920288,
      "learning_rate": 2.0740390674228104e-05,
      "loss": 1.8782,
      "step": 4000
    },
    {
      "epoch": 1.134032303768773,
      "eval_cosine_accuracy": 0.9714768528938293,
      "eval_loss": 0.8507260084152222,
      "eval_runtime": 27.9345,
      "eval_samples_per_second": 338.864,
      "eval_steps_per_second": 1.325,
      "step": 4000
    },
    {
      "epoch": 1.4173986965145935,
      "grad_norm": 4.125911712646484,
      "learning_rate": 1.7589792060491492e-05,
      "loss": 1.394,
      "step": 5000
    },
    {
      "epoch": 1.4173986965145935,
      "eval_cosine_accuracy": 0.9743291735649109,
      "eval_loss": 0.8180537223815918,
      "eval_runtime": 27.4596,
      "eval_samples_per_second": 344.725,
      "eval_steps_per_second": 1.347,
      "step": 5000
    },
    {
      "epoch": 1.7007650892604138,
      "grad_norm": 3.567063093185425,
      "learning_rate": 1.4439193446754883e-05,
      "loss": 1.3521,
      "step": 6000
    },
    {
      "epoch": 1.7007650892604138,
      "eval_cosine_accuracy": 0.9738009572029114,
      "eval_loss": 0.8179722428321838,
      "eval_runtime": 27.6707,
      "eval_samples_per_second": 342.095,
      "eval_steps_per_second": 1.337,
      "step": 6000
    },
    {
      "epoch": 1.9841314820062341,
      "grad_norm": 3.4355316162109375,
      "learning_rate": 1.1288594833018275e-05,
      "loss": 1.3137,
      "step": 7000
    },
    {
      "epoch": 1.9841314820062341,
      "eval_cosine_accuracy": 0.9742235541343689,
      "eval_loss": 0.8137649297714233,
      "eval_runtime": 27.3447,
      "eval_samples_per_second": 346.173,
      "eval_steps_per_second": 1.353,
      "step": 7000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 10581,
  "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
}