{
  "best_metric": 0.7557603686635944,
  "best_model_checkpoint": "/hps/software/users/chembl/christine/git_projects/OTAR3088/Entity-Extraction-Modular-pipeline/outputs/model_outputs/hf/model_outputs/hf/CeLLaTe_V2/reinit_llrd/4K_with_reinit_classifier_llrd0.9/bioformers/checkpoint-672",
  "epoch": 14.0,
  "eval_steps": 500,
  "global_step": 672,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 6.153050422668457,
      "learning_rate": 9.375000000000001e-06,
      "loss": 2.3155,
      "step": 48
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9688143571638718,
      "eval_f1": 0.0,
      "eval_loss": 0.2333633154630661,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 1.308,
      "eval_samples_per_second": 144.499,
      "eval_steps_per_second": 9.175,
      "step": 48
    },
    {
      "epoch": 2.0,
      "grad_norm": 7.159886360168457,
      "learning_rate": 1.9375e-05,
      "loss": 0.1837,
      "step": 96
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9791115033833481,
      "eval_f1": 0.37777777777777777,
      "eval_loss": 0.07473152875900269,
      "eval_precision": 0.4473684210526316,
      "eval_recall": 0.3269230769230769,
      "eval_runtime": 1.3056,
      "eval_samples_per_second": 144.764,
      "eval_steps_per_second": 9.191,
      "step": 96
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.33205464482307434,
      "learning_rate": 1.8958333333333334e-05,
      "loss": 0.0771,
      "step": 144
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9852897911150338,
      "eval_f1": 0.5420560747663552,
      "eval_loss": 0.056234169751405716,
      "eval_precision": 0.5272727272727272,
      "eval_recall": 0.5576923076923077,
      "eval_runtime": 1.3018,
      "eval_samples_per_second": 145.185,
      "eval_steps_per_second": 9.218,
      "step": 144
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.3881724178791046,
      "learning_rate": 1.7847222222222225e-05,
      "loss": 0.0532,
      "step": 192
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9872021182700794,
      "eval_f1": 0.6190476190476191,
      "eval_loss": 0.0470348596572876,
      "eval_precision": 0.6132075471698113,
      "eval_recall": 0.625,
      "eval_runtime": 1.3002,
      "eval_samples_per_second": 145.366,
      "eval_steps_per_second": 9.23,
      "step": 192
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.382257461547852,
      "learning_rate": 1.6736111111111113e-05,
      "loss": 0.0403,
      "step": 240
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9883789349808767,
      "eval_f1": 0.6415094339622641,
      "eval_loss": 0.04033330827951431,
      "eval_precision": 0.6296296296296297,
      "eval_recall": 0.6538461538461539,
      "eval_runtime": 1.3061,
      "eval_samples_per_second": 144.701,
      "eval_steps_per_second": 9.187,
      "step": 240
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.024486979469656944,
      "learning_rate": 1.5625e-05,
      "loss": 0.0308,
      "step": 288
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9877905266254781,
      "eval_f1": 0.6351931330472104,
      "eval_loss": 0.04146737605333328,
      "eval_precision": 0.5736434108527132,
      "eval_recall": 0.7115384615384616,
      "eval_runtime": 1.3002,
      "eval_samples_per_second": 145.36,
      "eval_steps_per_second": 9.229,
      "step": 288
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.022526560351252556,
      "learning_rate": 1.451388888888889e-05,
      "loss": 0.0234,
      "step": 336
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.989997057958223,
      "eval_f1": 0.7069767441860465,
      "eval_loss": 0.036325544118881226,
      "eval_precision": 0.6846846846846847,
      "eval_recall": 0.7307692307692307,
      "eval_runtime": 1.2996,
      "eval_samples_per_second": 145.425,
      "eval_steps_per_second": 9.233,
      "step": 336
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.0061901346780359745,
      "learning_rate": 1.3402777777777779e-05,
      "loss": 0.0183,
      "step": 384
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.990879670491321,
      "eval_f1": 0.6909090909090909,
      "eval_loss": 0.03637426719069481,
      "eval_precision": 0.6551724137931034,
      "eval_recall": 0.7307692307692307,
      "eval_runtime": 1.3096,
      "eval_samples_per_second": 144.324,
      "eval_steps_per_second": 9.163,
      "step": 384
    },
    {
      "epoch": 9.0,
      "grad_norm": 0.1458413451910019,
      "learning_rate": 1.2291666666666668e-05,
      "loss": 0.015,
      "step": 432
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9904383642247719,
      "eval_f1": 0.704225352112676,
      "eval_loss": 0.03453047573566437,
      "eval_precision": 0.6880733944954128,
      "eval_recall": 0.7211538461538461,
      "eval_runtime": 1.3147,
      "eval_samples_per_second": 143.761,
      "eval_steps_per_second": 9.128,
      "step": 432
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.05294337496161461,
      "learning_rate": 1.1180555555555557e-05,
      "loss": 0.0131,
      "step": 480
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.9905854663136217,
      "eval_f1": 0.7123287671232875,
      "eval_loss": 0.033786673098802567,
      "eval_precision": 0.6782608695652174,
      "eval_recall": 0.75,
      "eval_runtime": 1.3033,
      "eval_samples_per_second": 145.02,
      "eval_steps_per_second": 9.208,
      "step": 480
    },
    {
      "epoch": 11.0,
      "grad_norm": 0.021180154755711555,
      "learning_rate": 1.0069444444444445e-05,
      "loss": 0.0108,
      "step": 528
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.9904383642247719,
      "eval_f1": 0.6912442396313364,
      "eval_loss": 0.03413481265306473,
      "eval_precision": 0.6637168141592921,
      "eval_recall": 0.7211538461538461,
      "eval_runtime": 1.3026,
      "eval_samples_per_second": 145.096,
      "eval_steps_per_second": 9.212,
      "step": 528
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.045118171721696854,
      "learning_rate": 8.958333333333334e-06,
      "loss": 0.01,
      "step": 576
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.990879670491321,
      "eval_f1": 0.7264150943396226,
      "eval_loss": 0.03440742939710617,
      "eval_precision": 0.7129629629629629,
      "eval_recall": 0.7403846153846154,
      "eval_runtime": 1.3043,
      "eval_samples_per_second": 144.91,
      "eval_steps_per_second": 9.201,
      "step": 576
    },
    {
      "epoch": 13.0,
      "grad_norm": 1.282688856124878,
      "learning_rate": 7.847222222222223e-06,
      "loss": 0.0086,
      "step": 624
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.990879670491321,
      "eval_f1": 0.7348837209302327,
      "eval_loss": 0.035170141607522964,
      "eval_precision": 0.7117117117117117,
      "eval_recall": 0.7596153846153846,
      "eval_runtime": 1.3097,
      "eval_samples_per_second": 144.31,
      "eval_steps_per_second": 9.163,
      "step": 624
    },
    {
      "epoch": 14.0,
      "grad_norm": 0.04552546516060829,
      "learning_rate": 6.736111111111112e-06,
      "loss": 0.0072,
      "step": 672
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.9910267725801707,
      "eval_f1": 0.7557603686635944,
      "eval_loss": 0.034706298261880875,
      "eval_precision": 0.7256637168141593,
      "eval_recall": 0.7884615384615384,
      "eval_runtime": 1.3011,
      "eval_samples_per_second": 145.257,
      "eval_steps_per_second": 9.223,
      "step": 672
    }
  ],
  "logging_steps": 500,
  "max_steps": 960,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "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": 789505749171048.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}