{
  "best_metric": 0.020906077697873116,
  "best_model_checkpoint": "LLMNIDS-t5base-1/checkpoint-5043",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 5043,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "grad_norm": 6.331108093261719,
      "learning_rate": 1.6336633663366337e-05,
      "loss": 5.5187,
      "step": 170
    },
    {
      "epoch": 0.2,
      "grad_norm": 4.446777820587158,
      "learning_rate": 3.3168316831683175e-05,
      "loss": 0.8703,
      "step": 340
    },
    {
      "epoch": 0.3,
      "grad_norm": 3.2873497009277344,
      "learning_rate": 5e-05,
      "loss": 0.3904,
      "step": 510
    },
    {
      "epoch": 0.4,
      "grad_norm": 5.13348913192749,
      "learning_rate": 4.813794623182018e-05,
      "loss": 0.1922,
      "step": 680
    },
    {
      "epoch": 0.51,
      "grad_norm": 1.8617521524429321,
      "learning_rate": 4.626487439400617e-05,
      "loss": 0.1091,
      "step": 850
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.5739182829856873,
      "learning_rate": 4.439180255619216e-05,
      "loss": 0.0807,
      "step": 1020
    },
    {
      "epoch": 0.71,
      "grad_norm": 3.6188409328460693,
      "learning_rate": 4.251873071837814e-05,
      "loss": 0.0719,
      "step": 1190
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.20058074593544006,
      "learning_rate": 4.064565888056412e-05,
      "loss": 0.0449,
      "step": 1360
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.7106176614761353,
      "learning_rate": 3.877258704275011e-05,
      "loss": 0.0512,
      "step": 1530
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 4.5944,
      "eval_loss": 0.044217173010110855,
      "eval_rouge1": 96.4965,
      "eval_rouge2": 43.8853,
      "eval_rougeL": 96.5111,
      "eval_rougeLsum": 96.4892,
      "eval_runtime": 56.362,
      "eval_samples_per_second": 121.287,
      "eval_steps_per_second": 7.594,
      "step": 1681
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.03131448104977608,
      "learning_rate": 3.68995152049361e-05,
      "loss": 0.0372,
      "step": 1700
    },
    {
      "epoch": 1.11,
      "grad_norm": 0.7714371085166931,
      "learning_rate": 3.502644336712208e-05,
      "loss": 0.035,
      "step": 1870
    },
    {
      "epoch": 1.21,
      "grad_norm": 0.47366881370544434,
      "learning_rate": 3.316438959894227e-05,
      "loss": 0.0305,
      "step": 2040
    },
    {
      "epoch": 1.31,
      "grad_norm": 1.289427399635315,
      "learning_rate": 3.129131776112825e-05,
      "loss": 0.0295,
      "step": 2210
    },
    {
      "epoch": 1.42,
      "grad_norm": 1.7327721118927002,
      "learning_rate": 2.9440282062582637e-05,
      "loss": 0.0358,
      "step": 2380
    },
    {
      "epoch": 1.52,
      "grad_norm": 0.13644327223300934,
      "learning_rate": 2.756721022476862e-05,
      "loss": 0.0283,
      "step": 2550
    },
    {
      "epoch": 1.62,
      "grad_norm": 0.032617200165987015,
      "learning_rate": 2.5694138386954607e-05,
      "loss": 0.0213,
      "step": 2720
    },
    {
      "epoch": 1.72,
      "grad_norm": 0.21351170539855957,
      "learning_rate": 2.3821066549140592e-05,
      "loss": 0.0268,
      "step": 2890
    },
    {
      "epoch": 1.82,
      "grad_norm": 0.03627165034413338,
      "learning_rate": 2.1947994711326578e-05,
      "loss": 0.0196,
      "step": 3060
    },
    {
      "epoch": 1.92,
      "grad_norm": 0.08273334801197052,
      "learning_rate": 2.0074922873512563e-05,
      "loss": 0.0153,
      "step": 3230
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 4.559,
      "eval_loss": 0.026822404935956,
      "eval_rouge1": 98.2592,
      "eval_rouge2": 44.3827,
      "eval_rougeL": 98.2592,
      "eval_rougeLsum": 98.2665,
      "eval_runtime": 55.7197,
      "eval_samples_per_second": 122.686,
      "eval_steps_per_second": 7.681,
      "step": 3362
    },
    {
      "epoch": 2.02,
      "grad_norm": 0.009718131273984909,
      "learning_rate": 1.8201851035698548e-05,
      "loss": 0.0173,
      "step": 3400
    },
    {
      "epoch": 2.12,
      "grad_norm": 0.4162699282169342,
      "learning_rate": 1.632877919788453e-05,
      "loss": 0.0202,
      "step": 3570
    },
    {
      "epoch": 2.22,
      "grad_norm": 0.6483927965164185,
      "learning_rate": 1.4455707360070516e-05,
      "loss": 0.0166,
      "step": 3740
    },
    {
      "epoch": 2.33,
      "grad_norm": 0.029726264998316765,
      "learning_rate": 1.25826355222565e-05,
      "loss": 0.0138,
      "step": 3910
    },
    {
      "epoch": 2.43,
      "grad_norm": 0.008093880489468575,
      "learning_rate": 1.0709563684442487e-05,
      "loss": 0.0124,
      "step": 4080
    },
    {
      "epoch": 2.53,
      "grad_norm": 0.06707121431827545,
      "learning_rate": 8.836491846628472e-06,
      "loss": 0.0193,
      "step": 4250
    },
    {
      "epoch": 2.63,
      "grad_norm": 0.49983274936676025,
      "learning_rate": 6.963420008814456e-06,
      "loss": 0.0142,
      "step": 4420
    },
    {
      "epoch": 2.73,
      "grad_norm": 0.009119812399148941,
      "learning_rate": 5.090348171000441e-06,
      "loss": 0.0118,
      "step": 4590
    },
    {
      "epoch": 2.83,
      "grad_norm": 0.155787855386734,
      "learning_rate": 3.217276333186426e-06,
      "loss": 0.0235,
      "step": 4760
    },
    {
      "epoch": 2.93,
      "grad_norm": 0.08707331120967865,
      "learning_rate": 1.3442044953724108e-06,
      "loss": 0.0135,
      "step": 4930
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 4.5541,
      "eval_loss": 0.020906077697873116,
      "eval_rouge1": 98.3836,
      "eval_rouge2": 44.4266,
      "eval_rougeL": 98.3836,
      "eval_rougeLsum": 98.3909,
      "eval_runtime": 55.8601,
      "eval_samples_per_second": 122.377,
      "eval_steps_per_second": 7.662,
      "step": 5043
    }
  ],
  "logging_steps": 170,
  "max_steps": 5043,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 6138752255262720.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}