{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 2000,
  "global_step": 10863,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 9.205560158335634e-05,
      "grad_norm": 7.180713176727295,
      "learning_rate": 0.0,
      "loss": 2.5727,
      "step": 1
    },
    {
      "epoch": 0.1841112031667127,
      "grad_norm": 4.364831447601318,
      "learning_rate": 5.5210826735407845e-06,
      "loss": 2.2003,
      "step": 2000
    },
    {
      "epoch": 0.1841112031667127,
      "eval_cosine_accuracy": 0.9447852969169617,
      "eval_loss": 1.741607666015625,
      "eval_runtime": 20.9736,
      "eval_samples_per_second": 450.758,
      "eval_steps_per_second": 1.764,
      "step": 2000
    },
    {
      "epoch": 0.3682224063334254,
      "grad_norm": 4.162062644958496,
      "learning_rate": 1.1042165347081569e-05,
      "loss": 1.6012,
      "step": 4000
    },
    {
      "epoch": 0.3682224063334254,
      "eval_cosine_accuracy": 0.9465834498405457,
      "eval_loss": 1.685057520866394,
      "eval_runtime": 25.0673,
      "eval_samples_per_second": 377.145,
      "eval_steps_per_second": 1.476,
      "step": 4000
    },
    {
      "epoch": 0.5523336095001381,
      "grad_norm": 3.561060905456543,
      "learning_rate": 1.6563248020622354e-05,
      "loss": 1.2787,
      "step": 6000
    },
    {
      "epoch": 0.5523336095001381,
      "eval_cosine_accuracy": 0.9530357718467712,
      "eval_loss": 1.589287519454956,
      "eval_runtime": 23.3994,
      "eval_samples_per_second": 404.028,
      "eval_steps_per_second": 1.581,
      "step": 6000
    },
    {
      "epoch": 0.7364448126668508,
      "grad_norm": 3.908196449279785,
      "learning_rate": 2.2084330694163138e-05,
      "loss": 1.0753,
      "step": 8000
    },
    {
      "epoch": 0.7364448126668508,
      "eval_cosine_accuracy": 0.9574783444404602,
      "eval_loss": 1.5174190998077393,
      "eval_runtime": 23.6702,
      "eval_samples_per_second": 399.405,
      "eval_steps_per_second": 1.563,
      "step": 8000
    },
    {
      "epoch": 0.9205560158335635,
      "grad_norm": 3.514824628829956,
      "learning_rate": 2.760817529000184e-05,
      "loss": 0.9386,
      "step": 10000
    },
    {
      "epoch": 0.9205560158335635,
      "eval_cosine_accuracy": 0.9603342413902283,
      "eval_loss": 1.4767882823944092,
      "eval_runtime": 23.3417,
      "eval_samples_per_second": 405.026,
      "eval_steps_per_second": 1.585,
      "step": 10000
    }
  ],
  "logging_steps": 2000,
  "max_steps": 21726,
  "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
}