{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 1000,
  "global_step": 5515,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00018132366273798732,
      "grad_norm": 6.366542339324951,
      "learning_rate": 0.0,
      "loss": 1.2337,
      "step": 1
    },
    {
      "epoch": 0.1813236627379873,
      "grad_norm": 6.652426719665527,
      "learning_rate": 1.2072507552870091e-05,
      "loss": 1.2987,
      "step": 1000
    },
    {
      "epoch": 0.1813236627379873,
      "eval_cosine_accuracy": 0.9657166600227356,
      "eval_loss": 0.4266870319843292,
      "eval_runtime": 35.8127,
      "eval_samples_per_second": 265.52,
      "eval_steps_per_second": 2.094,
      "step": 1000
    },
    {
      "epoch": 0.3626473254759746,
      "grad_norm": 6.257086753845215,
      "learning_rate": 1.9539288112827402e-05,
      "loss": 1.2594,
      "step": 2000
    },
    {
      "epoch": 0.3626473254759746,
      "eval_cosine_accuracy": 0.9685561060905457,
      "eval_loss": 0.42614081501960754,
      "eval_runtime": 35.5879,
      "eval_samples_per_second": 267.198,
      "eval_steps_per_second": 2.107,
      "step": 2000
    },
    {
      "epoch": 0.543970988213962,
      "grad_norm": 6.342113494873047,
      "learning_rate": 1.8196104768300877e-05,
      "loss": 1.1928,
      "step": 3000
    },
    {
      "epoch": 0.543970988213962,
      "eval_cosine_accuracy": 0.9664528369903564,
      "eval_loss": 0.42114901542663574,
      "eval_runtime": 35.7842,
      "eval_samples_per_second": 265.732,
      "eval_steps_per_second": 2.096,
      "step": 3000
    },
    {
      "epoch": 0.7252946509519492,
      "grad_norm": 7.883905410766602,
      "learning_rate": 1.6852921423774348e-05,
      "loss": 0.9213,
      "step": 4000
    },
    {
      "epoch": 0.7252946509519492,
      "eval_cosine_accuracy": 0.9636133909225464,
      "eval_loss": 0.40882474184036255,
      "eval_runtime": 35.2731,
      "eval_samples_per_second": 269.582,
      "eval_steps_per_second": 2.126,
      "step": 4000
    },
    {
      "epoch": 0.9066183136899365,
      "grad_norm": 9.094715118408203,
      "learning_rate": 1.550973807924782e-05,
      "loss": 0.7272,
      "step": 5000
    },
    {
      "epoch": 0.9066183136899365,
      "eval_cosine_accuracy": 0.9598275423049927,
      "eval_loss": 0.41818034648895264,
      "eval_runtime": 35.7134,
      "eval_samples_per_second": 266.259,
      "eval_steps_per_second": 2.1,
      "step": 5000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 16545,
  "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": 128,
  "trial_name": null,
  "trial_params": null
}