{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 1000,
  "global_step": 14199,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00021128248468201986,
      "grad_norm": 7.594412803649902,
      "learning_rate": 0.0,
      "loss": 2.6278,
      "step": 1
    },
    {
      "epoch": 0.21128248468201985,
      "grad_norm": 6.050217628479004,
      "learning_rate": 8.441064638783271e-06,
      "loss": 2.7281,
      "step": 1000
    },
    {
      "epoch": 0.21128248468201985,
      "eval_cosine_accuracy": 0.9345883131027222,
      "eval_loss": 0.6125919222831726,
      "eval_runtime": 34.2392,
      "eval_samples_per_second": 277.723,
      "eval_steps_per_second": 2.19,
      "step": 1000
    },
    {
      "epoch": 0.4225649693640397,
      "grad_norm": 6.340444087982178,
      "learning_rate": 1.6890578791719477e-05,
      "loss": 1.999,
      "step": 2000
    },
    {
      "epoch": 0.4225649693640397,
      "eval_cosine_accuracy": 0.9388999938964844,
      "eval_loss": 0.5676162838935852,
      "eval_runtime": 34.1182,
      "eval_samples_per_second": 278.708,
      "eval_steps_per_second": 2.198,
      "step": 2000
    },
    {
      "epoch": 0.6338474540460596,
      "grad_norm": 8.218648910522461,
      "learning_rate": 1.940745609916424e-05,
      "loss": 1.5073,
      "step": 3000
    },
    {
      "epoch": 0.6338474540460596,
      "eval_cosine_accuracy": 0.9328005313873291,
      "eval_loss": 0.5519113540649414,
      "eval_runtime": 34.1651,
      "eval_samples_per_second": 278.325,
      "eval_steps_per_second": 2.195,
      "step": 3000
    },
    {
      "epoch": 0.8451299387280794,
      "grad_norm": 9.030098915100098,
      "learning_rate": 1.846933984411682e-05,
      "loss": 0.9687,
      "step": 4000
    },
    {
      "epoch": 0.8451299387280794,
      "eval_cosine_accuracy": 0.9312230348587036,
      "eval_loss": 0.5424668788909912,
      "eval_runtime": 34.165,
      "eval_samples_per_second": 278.326,
      "eval_steps_per_second": 2.195,
      "step": 4000
    },
    {
      "epoch": 1.0564124234100993,
      "grad_norm": 5.368319511413574,
      "learning_rate": 1.7530284533759038e-05,
      "loss": 0.9492,
      "step": 5000
    },
    {
      "epoch": 1.0564124234100993,
      "eval_cosine_accuracy": 0.944158136844635,
      "eval_loss": 0.5103689432144165,
      "eval_runtime": 33.9293,
      "eval_samples_per_second": 280.259,
      "eval_steps_per_second": 2.21,
      "step": 5000
    },
    {
      "epoch": 1.2676949080921192,
      "grad_norm": 6.123900890350342,
      "learning_rate": 1.6592168278711617e-05,
      "loss": 1.3563,
      "step": 6000
    },
    {
      "epoch": 1.2676949080921192,
      "eval_cosine_accuracy": 0.9431065320968628,
      "eval_loss": 0.516655445098877,
      "eval_runtime": 34.739,
      "eval_samples_per_second": 273.727,
      "eval_steps_per_second": 2.159,
      "step": 6000
    },
    {
      "epoch": 1.478977392774139,
      "grad_norm": 5.849966526031494,
      "learning_rate": 1.5653112968353836e-05,
      "loss": 1.253,
      "step": 7000
    },
    {
      "epoch": 1.478977392774139,
      "eval_cosine_accuracy": 0.9433168768882751,
      "eval_loss": 0.5244542956352234,
      "eval_runtime": 33.8791,
      "eval_samples_per_second": 280.674,
      "eval_steps_per_second": 2.214,
      "step": 7000
    },
    {
      "epoch": 1.690259877456159,
      "grad_norm": 8.606755256652832,
      "learning_rate": 1.4714057657996057e-05,
      "loss": 0.9613,
      "step": 8000
    },
    {
      "epoch": 1.690259877456159,
      "eval_cosine_accuracy": 0.9373225569725037,
      "eval_loss": 0.5143805742263794,
      "eval_runtime": 34.3167,
      "eval_samples_per_second": 277.096,
      "eval_steps_per_second": 2.186,
      "step": 8000
    },
    {
      "epoch": 1.9015423621381786,
      "grad_norm": 7.499396800994873,
      "learning_rate": 1.3775002347638278e-05,
      "loss": 0.6725,
      "step": 9000
    },
    {
      "epoch": 1.9015423621381786,
      "eval_cosine_accuracy": 0.9387948513031006,
      "eval_loss": 0.5081201791763306,
      "eval_runtime": 33.8383,
      "eval_samples_per_second": 281.013,
      "eval_steps_per_second": 2.216,
      "step": 9000
    },
    {
      "epoch": 2.1128248468201987,
      "grad_norm": 5.742690563201904,
      "learning_rate": 1.2835947037280495e-05,
      "loss": 0.8854,
      "step": 10000
    },
    {
      "epoch": 2.1128248468201987,
      "eval_cosine_accuracy": 0.944158136844635,
      "eval_loss": 0.49644434452056885,
      "eval_runtime": 33.7483,
      "eval_samples_per_second": 281.762,
      "eval_steps_per_second": 2.222,
      "step": 10000
    },
    {
      "epoch": 2.3241073315022183,
      "grad_norm": 5.786012172698975,
      "learning_rate": 1.1896891726922716e-05,
      "loss": 1.0927,
      "step": 11000
    },
    {
      "epoch": 2.3241073315022183,
      "eval_cosine_accuracy": 0.9468924403190613,
      "eval_loss": 0.49855560064315796,
      "eval_runtime": 34.3704,
      "eval_samples_per_second": 276.663,
      "eval_steps_per_second": 2.182,
      "step": 11000
    },
    {
      "epoch": 2.5353898161842383,
      "grad_norm": 5.753871917724609,
      "learning_rate": 1.0958775471875293e-05,
      "loss": 1.0451,
      "step": 12000
    },
    {
      "epoch": 2.5353898161842383,
      "eval_cosine_accuracy": 0.9464717507362366,
      "eval_loss": 0.48781055212020874,
      "eval_runtime": 34.0856,
      "eval_samples_per_second": 278.974,
      "eval_steps_per_second": 2.2,
      "step": 12000
    },
    {
      "epoch": 2.7466723008662584,
      "grad_norm": 8.100356101989746,
      "learning_rate": 1.0019720161517514e-05,
      "loss": 0.7421,
      "step": 13000
    },
    {
      "epoch": 2.7466723008662584,
      "eval_cosine_accuracy": 0.9420548677444458,
      "eval_loss": 0.48989957571029663,
      "eval_runtime": 34.0167,
      "eval_samples_per_second": 279.539,
      "eval_steps_per_second": 2.205,
      "step": 13000
    },
    {
      "epoch": 2.957954785548278,
      "grad_norm": 12.976683616638184,
      "learning_rate": 9.081603906470092e-06,
      "loss": 0.5394,
      "step": 14000
    },
    {
      "epoch": 2.957954785548278,
      "eval_cosine_accuracy": 0.9391103386878967,
      "eval_loss": 0.4943162500858307,
      "eval_runtime": 34.3547,
      "eval_samples_per_second": 276.789,
      "eval_steps_per_second": 2.183,
      "step": 14000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 23665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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
}