{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 2000,
  "global_step": 9497,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00010529640939243972,
      "grad_norm": 6.751945495605469,
      "learning_rate": 0.0,
      "loss": 2.8442,
      "step": 1
    },
    {
      "epoch": 0.21059281878487943,
      "grad_norm": 4.618383884429932,
      "learning_rate": 6.315290648694187e-06,
      "loss": 2.5015,
      "step": 2000
    },
    {
      "epoch": 0.21059281878487943,
      "eval_cosine_accuracy": 0.9469176530838013,
      "eval_loss": 0.7108581066131592,
      "eval_runtime": 24.9028,
      "eval_samples_per_second": 379.756,
      "eval_steps_per_second": 1.486,
      "step": 2000
    },
    {
      "epoch": 0.42118563756975885,
      "grad_norm": 4.100494861602783,
      "learning_rate": 1.2630581297388374e-05,
      "loss": 1.8564,
      "step": 4000
    },
    {
      "epoch": 0.42118563756975885,
      "eval_cosine_accuracy": 0.9481865167617798,
      "eval_loss": 0.6932559013366699,
      "eval_runtime": 25.2131,
      "eval_samples_per_second": 375.083,
      "eval_steps_per_second": 1.467,
      "step": 4000
    },
    {
      "epoch": 0.6317784563546383,
      "grad_norm": 4.006742477416992,
      "learning_rate": 1.8949031171019377e-05,
      "loss": 1.504,
      "step": 6000
    },
    {
      "epoch": 0.6317784563546383,
      "eval_cosine_accuracy": 0.9519932270050049,
      "eval_loss": 0.6632474064826965,
      "eval_runtime": 25.5415,
      "eval_samples_per_second": 370.261,
      "eval_steps_per_second": 1.449,
      "step": 6000
    },
    {
      "epoch": 0.8423712751395177,
      "grad_norm": 3.7652177810668945,
      "learning_rate": 2.5264321819713565e-05,
      "loss": 1.2874,
      "step": 8000
    },
    {
      "epoch": 0.8423712751395177,
      "eval_cosine_accuracy": 0.9577032923698425,
      "eval_loss": 0.6434259414672852,
      "eval_runtime": 25.6225,
      "eval_samples_per_second": 369.089,
      "eval_steps_per_second": 1.444,
      "step": 8000
    }
  ],
  "logging_steps": 2000,
  "max_steps": 18994,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "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
}