{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.757738896366083,
  "eval_steps": 500,
  "global_step": 14500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.34,
      "learning_rate": 1.932705248990579e-05,
      "loss": 0.2243,
      "step": 500
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.8654104979811575e-05,
      "loss": 0.1513,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.15557065606117249,
      "eval_runtime": 122.4795,
      "eval_samples_per_second": 12.141,
      "eval_steps_per_second": 3.037,
      "step": 1486
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.7981157469717363e-05,
      "loss": 0.1319,
      "step": 1500
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.730820995962315e-05,
      "loss": 0.1161,
      "step": 2000
    },
    {
      "epoch": 1.68,
      "learning_rate": 1.6635262449528936e-05,
      "loss": 0.1161,
      "step": 2500
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.09562207758426666,
      "eval_runtime": 122.4632,
      "eval_samples_per_second": 12.142,
      "eval_steps_per_second": 3.038,
      "step": 2972
    },
    {
      "epoch": 2.02,
      "learning_rate": 1.5962314939434724e-05,
      "loss": 0.1072,
      "step": 3000
    },
    {
      "epoch": 2.36,
      "learning_rate": 1.5289367429340512e-05,
      "loss": 0.0999,
      "step": 3500
    },
    {
      "epoch": 2.69,
      "learning_rate": 1.46164199192463e-05,
      "loss": 0.0765,
      "step": 4000
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.11794405430555344,
      "eval_runtime": 122.4644,
      "eval_samples_per_second": 12.142,
      "eval_steps_per_second": 3.038,
      "step": 4458
    },
    {
      "epoch": 3.03,
      "learning_rate": 1.3943472409152087e-05,
      "loss": 0.0886,
      "step": 4500
    },
    {
      "epoch": 3.36,
      "learning_rate": 1.3270524899057875e-05,
      "loss": 0.0793,
      "step": 5000
    },
    {
      "epoch": 3.7,
      "learning_rate": 1.2597577388963662e-05,
      "loss": 0.0798,
      "step": 5500
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.11743835359811783,
      "eval_runtime": 122.4332,
      "eval_samples_per_second": 12.145,
      "eval_steps_per_second": 3.038,
      "step": 5944
    },
    {
      "epoch": 4.04,
      "learning_rate": 1.192462987886945e-05,
      "loss": 0.0737,
      "step": 6000
    },
    {
      "epoch": 4.37,
      "learning_rate": 1.1251682368775237e-05,
      "loss": 0.0739,
      "step": 6500
    },
    {
      "epoch": 4.71,
      "learning_rate": 1.0578734858681023e-05,
      "loss": 0.0705,
      "step": 7000
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.09409211575984955,
      "eval_runtime": 122.4407,
      "eval_samples_per_second": 12.145,
      "eval_steps_per_second": 3.038,
      "step": 7430
    },
    {
      "epoch": 5.05,
      "learning_rate": 9.905787348586811e-06,
      "loss": 0.0668,
      "step": 7500
    },
    {
      "epoch": 5.38,
      "learning_rate": 9.2328398384926e-06,
      "loss": 0.0558,
      "step": 8000
    },
    {
      "epoch": 5.72,
      "learning_rate": 8.559892328398386e-06,
      "loss": 0.0609,
      "step": 8500
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.12117910385131836,
      "eval_runtime": 122.4449,
      "eval_samples_per_second": 12.144,
      "eval_steps_per_second": 3.038,
      "step": 8916
    },
    {
      "epoch": 6.06,
      "learning_rate": 7.886944818304172e-06,
      "loss": 0.0617,
      "step": 9000
    },
    {
      "epoch": 6.39,
      "learning_rate": 7.21399730820996e-06,
      "loss": 0.0537,
      "step": 9500
    },
    {
      "epoch": 6.73,
      "learning_rate": 6.541049798115747e-06,
      "loss": 0.0542,
      "step": 10000
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.10626289248466492,
      "eval_runtime": 122.4232,
      "eval_samples_per_second": 12.146,
      "eval_steps_per_second": 3.039,
      "step": 10402
    },
    {
      "epoch": 7.07,
      "learning_rate": 5.8681022880215346e-06,
      "loss": 0.0485,
      "step": 10500
    },
    {
      "epoch": 7.4,
      "learning_rate": 5.195154777927323e-06,
      "loss": 0.0522,
      "step": 11000
    },
    {
      "epoch": 7.74,
      "learning_rate": 4.522207267833109e-06,
      "loss": 0.0542,
      "step": 11500
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.13295996189117432,
      "eval_runtime": 122.4378,
      "eval_samples_per_second": 12.145,
      "eval_steps_per_second": 3.038,
      "step": 11888
    },
    {
      "epoch": 8.08,
      "learning_rate": 3.849259757738897e-06,
      "loss": 0.0545,
      "step": 12000
    },
    {
      "epoch": 8.41,
      "learning_rate": 3.176312247644684e-06,
      "loss": 0.0391,
      "step": 12500
    },
    {
      "epoch": 8.75,
      "learning_rate": 2.503364737550471e-06,
      "loss": 0.0418,
      "step": 13000
    },
    {
      "epoch": 9.0,
      "eval_loss": 0.1395300030708313,
      "eval_runtime": 122.4262,
      "eval_samples_per_second": 12.146,
      "eval_steps_per_second": 3.039,
      "step": 13374
    },
    {
      "epoch": 9.08,
      "learning_rate": 1.8304172274562585e-06,
      "loss": 0.0385,
      "step": 13500
    },
    {
      "epoch": 9.42,
      "learning_rate": 1.157469717362046e-06,
      "loss": 0.0434,
      "step": 14000
    },
    {
      "epoch": 9.76,
      "learning_rate": 4.845222072678332e-07,
      "loss": 0.0494,
      "step": 14500
    }
  ],
  "logging_steps": 500,
  "max_steps": 14860,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 2.61086994432e+17,
  "trial_name": null,
  "trial_params": null
}