{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.88997555012225,
  "eval_steps": 800,
  "global_step": 18000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.22,
      "learning_rate": 4.7826677533279e-05,
      "loss": 0.3854,
      "step": 800
    },
    {
      "epoch": 0.22,
      "eval_accuracy": 0.8774075206358911,
      "eval_f1-score": 0.8772242175124827,
      "eval_loss": 0.3071616590023041,
      "eval_precision": 0.8794550305873866,
      "eval_recall": 0.8774075206358911,
      "eval_runtime": 64.5762,
      "eval_samples_per_second": 50.653,
      "eval_steps_per_second": 6.334,
      "step": 800
    },
    {
      "epoch": 0.43,
      "learning_rate": 4.5653355066558e-05,
      "loss": 0.3445,
      "step": 1600
    },
    {
      "epoch": 0.43,
      "eval_accuracy": 0.9003362885967594,
      "eval_f1-score": 0.9003365494138151,
      "eval_loss": 0.27588778734207153,
      "eval_precision": 0.9003395029663702,
      "eval_recall": 0.9003362885967594,
      "eval_runtime": 60.4968,
      "eval_samples_per_second": 54.069,
      "eval_steps_per_second": 6.761,
      "step": 1600
    },
    {
      "epoch": 0.65,
      "learning_rate": 4.3480032599837e-05,
      "loss": 0.3277,
      "step": 2400
    },
    {
      "epoch": 0.65,
      "eval_accuracy": 0.8969734026291654,
      "eval_f1-score": 0.8969608633994076,
      "eval_loss": 0.288944274187088,
      "eval_precision": 0.8971095244147143,
      "eval_recall": 0.8969734026291654,
      "eval_runtime": 60.3986,
      "eval_samples_per_second": 54.157,
      "eval_steps_per_second": 6.772,
      "step": 2400
    },
    {
      "epoch": 0.87,
      "learning_rate": 4.130671013311601e-05,
      "loss": 0.3222,
      "step": 3200
    },
    {
      "epoch": 0.87,
      "eval_accuracy": 0.9046163252827881,
      "eval_f1-score": 0.9045643986899728,
      "eval_loss": 0.3288567364215851,
      "eval_precision": 0.9056517879747317,
      "eval_recall": 0.9046163252827881,
      "eval_runtime": 60.7827,
      "eval_samples_per_second": 53.815,
      "eval_steps_per_second": 6.729,
      "step": 3200
    },
    {
      "epoch": 1.09,
      "learning_rate": 3.9133387666395004e-05,
      "loss": 0.2717,
      "step": 4000
    },
    {
      "epoch": 1.09,
      "eval_accuracy": 0.9021705900336289,
      "eval_f1-score": 0.9020992071629579,
      "eval_loss": 0.42761483788490295,
      "eval_precision": 0.9035219658491396,
      "eval_recall": 0.9021705900336289,
      "eval_runtime": 62.9323,
      "eval_samples_per_second": 51.976,
      "eval_steps_per_second": 6.499,
      "step": 4000
    },
    {
      "epoch": 1.3,
      "learning_rate": 3.696006519967401e-05,
      "loss": 0.2167,
      "step": 4800
    },
    {
      "epoch": 1.3,
      "eval_accuracy": 0.908590645062672,
      "eval_f1-score": 0.9085664450086591,
      "eval_loss": 0.4045434892177582,
      "eval_precision": 0.9091340460574109,
      "eval_recall": 0.908590645062672,
      "eval_runtime": 61.45,
      "eval_samples_per_second": 53.23,
      "eval_steps_per_second": 6.656,
      "step": 4800
    },
    {
      "epoch": 1.52,
      "learning_rate": 3.4786742732953e-05,
      "loss": 0.2092,
      "step": 5600
    },
    {
      "epoch": 1.52,
      "eval_accuracy": 0.9116478141241211,
      "eval_f1-score": 0.9116147566351114,
      "eval_loss": 0.3573218584060669,
      "eval_precision": 0.9123941687742838,
      "eval_recall": 0.9116478141241211,
      "eval_runtime": 62.314,
      "eval_samples_per_second": 52.492,
      "eval_steps_per_second": 6.564,
      "step": 5600
    },
    {
      "epoch": 1.74,
      "learning_rate": 3.2613420266232006e-05,
      "loss": 0.2062,
      "step": 6400
    },
    {
      "epoch": 1.74,
      "eval_accuracy": 0.904922042188933,
      "eval_f1-score": 0.9049222910044366,
      "eval_loss": 0.3768712878227234,
      "eval_precision": 0.9049308283306433,
      "eval_recall": 0.904922042188933,
      "eval_runtime": 63.1267,
      "eval_samples_per_second": 51.816,
      "eval_steps_per_second": 6.479,
      "step": 6400
    },
    {
      "epoch": 1.96,
      "learning_rate": 3.0440097799511003e-05,
      "loss": 0.2254,
      "step": 7200
    },
    {
      "epoch": 1.96,
      "eval_accuracy": 0.908590645062672,
      "eval_f1-score": 0.9084747894886354,
      "eval_loss": 0.4190152585506439,
      "eval_precision": 0.910461333213157,
      "eval_recall": 0.908590645062672,
      "eval_runtime": 61.865,
      "eval_samples_per_second": 52.873,
      "eval_steps_per_second": 6.611,
      "step": 7200
    },
    {
      "epoch": 2.17,
      "learning_rate": 2.8266775332790002e-05,
      "loss": 0.1289,
      "step": 8000
    },
    {
      "epoch": 2.17,
      "eval_accuracy": 0.8994191378783247,
      "eval_f1-score": 0.8992916579992009,
      "eval_loss": 0.5542723536491394,
      "eval_precision": 0.9012452606359794,
      "eval_recall": 0.8994191378783247,
      "eval_runtime": 62.1736,
      "eval_samples_per_second": 52.611,
      "eval_steps_per_second": 6.578,
      "step": 8000
    },
    {
      "epoch": 2.39,
      "learning_rate": 2.6093452866069005e-05,
      "loss": 0.1046,
      "step": 8800
    },
    {
      "epoch": 2.39,
      "eval_accuracy": 0.9134821155609906,
      "eval_f1-score": 0.9134820185252122,
      "eval_loss": 0.4697829782962799,
      "eval_precision": 0.9134820922449185,
      "eval_recall": 0.9134821155609906,
      "eval_runtime": 62.1014,
      "eval_samples_per_second": 52.672,
      "eval_steps_per_second": 6.586,
      "step": 8800
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.3920130399348005e-05,
      "loss": 0.1257,
      "step": 9600
    },
    {
      "epoch": 2.61,
      "eval_accuracy": 0.9131763986548456,
      "eval_f1-score": 0.9131764798026453,
      "eval_loss": 0.4252018332481384,
      "eval_precision": 0.9131936306903841,
      "eval_recall": 0.9131763986548456,
      "eval_runtime": 63.346,
      "eval_samples_per_second": 51.637,
      "eval_steps_per_second": 6.457,
      "step": 9600
    },
    {
      "epoch": 2.83,
      "learning_rate": 2.1746807932627005e-05,
      "loss": 0.1132,
      "step": 10400
    },
    {
      "epoch": 2.83,
      "eval_accuracy": 0.9131763986548456,
      "eval_f1-score": 0.9131652307887081,
      "eval_loss": 0.42821380496025085,
      "eval_precision": 0.9133289397789441,
      "eval_recall": 0.9131763986548456,
      "eval_runtime": 63.3473,
      "eval_samples_per_second": 51.636,
      "eval_steps_per_second": 6.456,
      "step": 10400
    },
    {
      "epoch": 3.04,
      "learning_rate": 1.9573485465906004e-05,
      "loss": 0.0964,
      "step": 11200
    },
    {
      "epoch": 3.04,
      "eval_accuracy": 0.9162335677162947,
      "eval_f1-score": 0.9162321584770682,
      "eval_loss": 0.5392835140228271,
      "eval_precision": 0.9162992614461799,
      "eval_recall": 0.9162335677162947,
      "eval_runtime": 61.748,
      "eval_samples_per_second": 52.973,
      "eval_steps_per_second": 6.624,
      "step": 11200
    },
    {
      "epoch": 3.26,
      "learning_rate": 1.7400162999185004e-05,
      "loss": 0.0476,
      "step": 12000
    },
    {
      "epoch": 3.26,
      "eval_accuracy": 0.9177621522470193,
      "eval_f1-score": 0.9177552338454668,
      "eval_loss": 0.5539752244949341,
      "eval_precision": 0.9178555704146932,
      "eval_recall": 0.9177621522470193,
      "eval_runtime": 62.8946,
      "eval_samples_per_second": 52.008,
      "eval_steps_per_second": 6.503,
      "step": 12000
    },
    {
      "epoch": 3.48,
      "learning_rate": 1.5226840532464004e-05,
      "loss": 0.0436,
      "step": 12800
    },
    {
      "epoch": 3.48,
      "eval_accuracy": 0.9128706817487007,
      "eval_f1-score": 0.9128070948010026,
      "eval_loss": 0.5679276585578918,
      "eval_precision": 0.9139222236120726,
      "eval_recall": 0.9128706817487007,
      "eval_runtime": 62.7176,
      "eval_samples_per_second": 52.154,
      "eval_steps_per_second": 6.521,
      "step": 12800
    },
    {
      "epoch": 3.69,
      "learning_rate": 1.3053518065743007e-05,
      "loss": 0.0582,
      "step": 13600
    },
    {
      "epoch": 3.69,
      "eval_accuracy": 0.9153164169978599,
      "eval_f1-score": 0.9152738804862125,
      "eval_loss": 0.5282616019248962,
      "eval_precision": 0.9163011804874139,
      "eval_recall": 0.9153164169978599,
      "eval_runtime": 62.1777,
      "eval_samples_per_second": 52.607,
      "eval_steps_per_second": 6.578,
      "step": 13600
    },
    {
      "epoch": 3.91,
      "learning_rate": 1.0880195599022006e-05,
      "loss": 0.0478,
      "step": 14400
    },
    {
      "epoch": 3.91,
      "eval_accuracy": 0.9159278508101498,
      "eval_f1-score": 0.9159151975384024,
      "eval_loss": 0.5375821590423584,
      "eval_precision": 0.9161123951905313,
      "eval_recall": 0.9159278508101498,
      "eval_runtime": 64.9703,
      "eval_samples_per_second": 50.346,
      "eval_steps_per_second": 6.295,
      "step": 14400
    },
    {
      "epoch": 4.13,
      "learning_rate": 8.706873132301006e-06,
      "loss": 0.0379,
      "step": 15200
    },
    {
      "epoch": 4.13,
      "eval_accuracy": 0.9159278508101498,
      "eval_f1-score": 0.9158993160297756,
      "eval_loss": 0.5539520978927612,
      "eval_precision": 0.916389347487138,
      "eval_recall": 0.9159278508101498,
      "eval_runtime": 61.7598,
      "eval_samples_per_second": 52.963,
      "eval_steps_per_second": 6.622,
      "step": 15200
    },
    {
      "epoch": 4.35,
      "learning_rate": 6.5335506655800055e-06,
      "loss": 0.0198,
      "step": 16000
    },
    {
      "epoch": 4.35,
      "eval_accuracy": 0.9143992662794252,
      "eval_f1-score": 0.9143978261809456,
      "eval_loss": 0.6111006140708923,
      "eval_precision": 0.9144647611244408,
      "eval_recall": 0.9143992662794252,
      "eval_runtime": 63.6411,
      "eval_samples_per_second": 51.398,
      "eval_steps_per_second": 6.427,
      "step": 16000
    },
    {
      "epoch": 4.56,
      "learning_rate": 4.360228198859006e-06,
      "loss": 0.0195,
      "step": 16800
    },
    {
      "epoch": 4.56,
      "eval_accuracy": 0.9159278508101498,
      "eval_f1-score": 0.915924110308607,
      "eval_loss": 0.5844493508338928,
      "eval_precision": 0.9159698581146932,
      "eval_recall": 0.9159278508101498,
      "eval_runtime": 63.4625,
      "eval_samples_per_second": 51.542,
      "eval_steps_per_second": 6.445,
      "step": 16800
    },
    {
      "epoch": 4.78,
      "learning_rate": 2.186905732138006e-06,
      "loss": 0.0189,
      "step": 17600
    },
    {
      "epoch": 4.78,
      "eval_accuracy": 0.9174564353408744,
      "eval_f1-score": 0.9174563427638073,
      "eval_loss": 0.5772590637207031,
      "eval_precision": 0.917480946018594,
      "eval_recall": 0.9174564353408744,
      "eval_runtime": 62.6716,
      "eval_samples_per_second": 52.193,
      "eval_steps_per_second": 6.526,
      "step": 17600
    }
  ],
  "logging_steps": 800,
  "max_steps": 18405,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 3.788201743080653e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}