{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 1000,
  "global_step": 10154,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00039416633819471815,
      "grad_norm": 6.560182094573975,
      "learning_rate": 0.0,
      "loss": 3.906,
      "step": 1
    },
    {
      "epoch": 0.3941663381947182,
      "grad_norm": 4.686622619628906,
      "learning_rate": 1.3118844386080105e-05,
      "loss": 3.3873,
      "step": 1000
    },
    {
      "epoch": 0.3941663381947182,
      "eval_cosine_accuracy": 0.9577242732048035,
      "eval_loss": 1.444562554359436,
      "eval_runtime": 24.0999,
      "eval_samples_per_second": 394.567,
      "eval_steps_per_second": 1.577,
      "step": 1000
    },
    {
      "epoch": 0.7883326763894364,
      "grad_norm": 4.804063320159912,
      "learning_rate": 1.9305058763413388e-05,
      "loss": 2.6104,
      "step": 2000
    },
    {
      "epoch": 0.7883326763894364,
      "eval_cosine_accuracy": 0.9649805426597595,
      "eval_loss": 1.3182923793792725,
      "eval_runtime": 24.3741,
      "eval_samples_per_second": 390.127,
      "eval_steps_per_second": 1.559,
      "step": 2000
    },
    {
      "epoch": 1.182355257975581,
      "grad_norm": 4.845874786376953,
      "learning_rate": 1.7845098182349078e-05,
      "loss": 1.9638,
      "step": 3000
    },
    {
      "epoch": 1.182355257975581,
      "eval_cosine_accuracy": 0.9649805426597595,
      "eval_loss": 1.2991231679916382,
      "eval_runtime": 24.0097,
      "eval_samples_per_second": 396.049,
      "eval_steps_per_second": 1.583,
      "step": 3000
    },
    {
      "epoch": 1.576211106734935,
      "grad_norm": 4.727691173553467,
      "learning_rate": 1.6386597561865833e-05,
      "loss": 1.9904,
      "step": 4000
    },
    {
      "epoch": 1.576211106734935,
      "eval_cosine_accuracy": 0.9685561060905457,
      "eval_loss": 1.2640169858932495,
      "eval_runtime": 23.8211,
      "eval_samples_per_second": 399.185,
      "eval_steps_per_second": 1.595,
      "step": 4000
    },
    {
      "epoch": 1.9700669554942891,
      "grad_norm": 4.540011405944824,
      "learning_rate": 1.4926636980801519e-05,
      "loss": 1.8765,
      "step": 5000
    },
    {
      "epoch": 1.9700669554942891,
      "eval_cosine_accuracy": 0.9704490303993225,
      "eval_loss": 1.2718477249145508,
      "eval_runtime": 23.8069,
      "eval_samples_per_second": 399.422,
      "eval_steps_per_second": 1.596,
      "step": 5000
    },
    {
      "epoch": 2.3639228042536433,
      "grad_norm": 9.497809410095215,
      "learning_rate": 1.3466676399737208e-05,
      "loss": 1.7538,
      "step": 6000
    },
    {
      "epoch": 2.3639228042536433,
      "eval_cosine_accuracy": 0.9702387452125549,
      "eval_loss": 1.2771531343460083,
      "eval_runtime": 25.1947,
      "eval_samples_per_second": 377.421,
      "eval_steps_per_second": 1.508,
      "step": 6000
    },
    {
      "epoch": 2.7577786530129975,
      "grad_norm": 4.704818248748779,
      "learning_rate": 1.2006715818672896e-05,
      "loss": 1.6971,
      "step": 7000
    },
    {
      "epoch": 2.7577786530129975,
      "eval_cosine_accuracy": 0.9713954925537109,
      "eval_loss": 1.2673554420471191,
      "eval_runtime": 23.8332,
      "eval_samples_per_second": 398.981,
      "eval_steps_per_second": 1.594,
      "step": 7000
    },
    {
      "epoch": 3.1516345017723513,
      "grad_norm": 4.666325569152832,
      "learning_rate": 1.0548215198189651e-05,
      "loss": 1.6378,
      "step": 8000
    },
    {
      "epoch": 3.1516345017723513,
      "eval_cosine_accuracy": 0.9723420143127441,
      "eval_loss": 1.2950376272201538,
      "eval_runtime": 23.8005,
      "eval_samples_per_second": 399.53,
      "eval_steps_per_second": 1.597,
      "step": 8000
    },
    {
      "epoch": 3.5454903505317055,
      "grad_norm": 4.480121612548828,
      "learning_rate": 9.088254617125339e-06,
      "loss": 1.5757,
      "step": 9000
    },
    {
      "epoch": 3.5454903505317055,
      "eval_cosine_accuracy": 0.9733936190605164,
      "eval_loss": 1.3058724403381348,
      "eval_runtime": 23.7962,
      "eval_samples_per_second": 399.602,
      "eval_steps_per_second": 1.597,
      "step": 9000
    },
    {
      "epoch": 3.9393461992910597,
      "grad_norm": 4.783756256103516,
      "learning_rate": 7.62975399664209e-06,
      "loss": 1.5623,
      "step": 10000
    },
    {
      "epoch": 3.9393461992910597,
      "eval_cosine_accuracy": 0.9734987616539001,
      "eval_loss": 1.2862597703933716,
      "eval_runtime": 23.7755,
      "eval_samples_per_second": 399.95,
      "eval_steps_per_second": 1.598,
      "step": 10000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 15222,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "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
}