{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 1000,
  "global_step": 10148,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00039416633819471815,
      "grad_norm": 6.644246578216553,
      "learning_rate": 0.0,
      "loss": 4.1042,
      "step": 1
    },
    {
      "epoch": 0.3941663381947182,
      "grad_norm": 4.322690963745117,
      "learning_rate": 3.276428102429416e-05,
      "loss": 3.3834,
      "step": 1000
    },
    {
      "epoch": 0.3941663381947182,
      "eval_cosine_accuracy": 0.9609843492507935,
      "eval_loss": 1.7424726486206055,
      "eval_runtime": 22.1887,
      "eval_samples_per_second": 428.552,
      "eval_steps_per_second": 1.713,
      "step": 1000
    },
    {
      "epoch": 0.7883326763894364,
      "grad_norm": 4.4025044441223145,
      "learning_rate": 4.82699467114388e-05,
      "loss": 2.5243,
      "step": 2000
    },
    {
      "epoch": 0.7883326763894364,
      "eval_cosine_accuracy": 0.967714786529541,
      "eval_loss": 1.6491715908050537,
      "eval_runtime": 22.146,
      "eval_samples_per_second": 429.378,
      "eval_steps_per_second": 1.716,
      "step": 2000
    },
    {
      "epoch": 1.1824990145841545,
      "grad_norm": 4.559783458709717,
      "learning_rate": 4.462004525877802e-05,
      "loss": 2.1493,
      "step": 3000
    },
    {
      "epoch": 1.1824990145841545,
      "eval_cosine_accuracy": 0.9698180556297302,
      "eval_loss": 1.6139544248580933,
      "eval_runtime": 22.1217,
      "eval_samples_per_second": 429.85,
      "eval_steps_per_second": 1.718,
      "step": 3000
    },
    {
      "epoch": 1.5766653527788725,
      "grad_norm": 4.444276809692383,
      "learning_rate": 4.097014380611724e-05,
      "loss": 1.9646,
      "step": 4000
    },
    {
      "epoch": 1.5766653527788725,
      "eval_cosine_accuracy": 0.9717110395431519,
      "eval_loss": 1.5907618999481201,
      "eval_runtime": 22.1029,
      "eval_samples_per_second": 430.216,
      "eval_steps_per_second": 1.719,
      "step": 4000
    },
    {
      "epoch": 1.9708316909735908,
      "grad_norm": 3.9484176635742188,
      "learning_rate": 3.732024235345646e-05,
      "loss": 1.8568,
      "step": 5000
    },
    {
      "epoch": 1.9708316909735908,
      "eval_cosine_accuracy": 0.9738143086433411,
      "eval_loss": 1.577061653137207,
      "eval_runtime": 22.108,
      "eval_samples_per_second": 430.115,
      "eval_steps_per_second": 1.719,
      "step": 5000
    },
    {
      "epoch": 2.364998029168309,
      "grad_norm": 3.979762554168701,
      "learning_rate": 3.3670340900795685e-05,
      "loss": 1.732,
      "step": 6000
    },
    {
      "epoch": 2.364998029168309,
      "eval_cosine_accuracy": 0.9728677868843079,
      "eval_loss": 1.604906439781189,
      "eval_runtime": 23.0465,
      "eval_samples_per_second": 412.601,
      "eval_steps_per_second": 1.649,
      "step": 6000
    },
    {
      "epoch": 2.7591643673630273,
      "grad_norm": 4.112518310546875,
      "learning_rate": 3.00204394481349e-05,
      "loss": 1.6763,
      "step": 7000
    },
    {
      "epoch": 2.7591643673630273,
      "eval_cosine_accuracy": 0.9752865433692932,
      "eval_loss": 1.5851424932479858,
      "eval_runtime": 22.0421,
      "eval_samples_per_second": 431.403,
      "eval_steps_per_second": 1.724,
      "step": 7000
    },
    {
      "epoch": 3.1533307055577455,
      "grad_norm": 4.042572975158691,
      "learning_rate": 2.6370537995474125e-05,
      "loss": 1.6198,
      "step": 8000
    },
    {
      "epoch": 3.1533307055577455,
      "eval_cosine_accuracy": 0.9754968881607056,
      "eval_loss": 1.604630470275879,
      "eval_runtime": 22.1791,
      "eval_samples_per_second": 428.737,
      "eval_steps_per_second": 1.713,
      "step": 8000
    },
    {
      "epoch": 3.5474970437524638,
      "grad_norm": 3.7061877250671387,
      "learning_rate": 2.2720636542813345e-05,
      "loss": 1.5682,
      "step": 9000
    },
    {
      "epoch": 3.5474970437524638,
      "eval_cosine_accuracy": 0.9758123755455017,
      "eval_loss": 1.600504994392395,
      "eval_runtime": 22.0036,
      "eval_samples_per_second": 432.156,
      "eval_steps_per_second": 1.727,
      "step": 9000
    },
    {
      "epoch": 3.9416633819471816,
      "grad_norm": 4.005512714385986,
      "learning_rate": 1.9070735090152565e-05,
      "loss": 1.5469,
      "step": 10000
    },
    {
      "epoch": 3.9416633819471816,
      "eval_cosine_accuracy": 0.9752865433692932,
      "eval_loss": 1.608902931213379,
      "eval_runtime": 22.1524,
      "eval_samples_per_second": 429.255,
      "eval_steps_per_second": 1.715,
      "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
}