trainer_state.json

{
  "best_metric": 0.32368871569633484,
  "best_model_checkpoint": "gal_mBERT/checkpoint-1250",
  "epoch": 6.0,
  "global_step": 1500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8816548086359176,
      "eval_f1": 0.5462899150011488,
      "eval_loss": 0.42043304443359375,
      "eval_precision": 0.5816076960704386,
      "eval_recall": 0.5150158821830783,
      "eval_runtime": 5.2868,
      "eval_samples_per_second": 188.582,
      "eval_steps_per_second": 11.916,
      "step": 250
    },
    {
      "epoch": 2.0,
      "learning_rate": 1.9600000000000002e-05,
      "loss": 0.5392,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8960377821393523,
      "eval_f1": 0.6141933543827995,
      "eval_loss": 0.35990580916404724,
      "eval_precision": 0.6406838143036386,
      "eval_recall": 0.5898065261334103,
      "eval_runtime": 5.3036,
      "eval_samples_per_second": 187.984,
      "eval_steps_per_second": 11.879,
      "step": 500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9000245338567223,
      "eval_f1": 0.6498169529709941,
      "eval_loss": 0.3427206575870514,
      "eval_precision": 0.6341027754877714,
      "eval_recall": 0.6663297718740976,
      "eval_runtime": 5.3775,
      "eval_samples_per_second": 185.401,
      "eval_steps_per_second": 11.715,
      "step": 750
    },
    {
      "epoch": 4.0,
      "learning_rate": 1.9200000000000003e-05,
      "loss": 0.2739,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.907262021589794,
      "eval_f1": 0.6595900439238652,
      "eval_loss": 0.3309294283390045,
      "eval_precision": 0.668993168993169,
      "eval_recall": 0.6504475887958417,
      "eval_runtime": 5.2835,
      "eval_samples_per_second": 188.7,
      "eval_steps_per_second": 11.924,
      "step": 1000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9131194798822375,
      "eval_f1": 0.6891636363636363,
      "eval_loss": 0.32368871569633484,
      "eval_precision": 0.6943141852286049,
      "eval_recall": 0.6840889402252383,
      "eval_runtime": 5.2998,
      "eval_samples_per_second": 188.119,
      "eval_steps_per_second": 11.887,
      "step": 1250
    },
    {
      "epoch": 6.0,
      "learning_rate": 1.88e-05,
      "loss": 0.1948,
      "step": 1500
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9126288027477919,
      "eval_f1": 0.69868810833686,
      "eval_loss": 0.3336099088191986,
      "eval_precision": 0.6829840044125759,
      "eval_recall": 0.7151313889691019,
      "eval_runtime": 5.4071,
      "eval_samples_per_second": 184.388,
      "eval_steps_per_second": 11.651,
      "step": 1500
    }
  ],
  "max_steps": 25000,
  "num_train_epochs": 100,
  "total_flos": 934904811170400.0,
  "trial_name": null,
  "trial_params": null
}