trainer_state.json

{
  "best_metric": 0.1305142641067505,
  "best_model_checkpoint": "/data/jcanete/all_results/mldoc/beto_uncased/epochs_3_bs_32_lr_2e-5/checkpoint-700",
  "epoch": 3.0,
  "global_step": 888,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.34,
      "eval_accuracy": 0.9110000133514404,
      "eval_loss": 0.33802908658981323,
      "eval_runtime": 8.5211,
      "eval_samples_per_second": 117.356,
      "eval_steps_per_second": 3.755,
      "step": 100
    },
    {
      "epoch": 0.68,
      "eval_accuracy": 0.921999990940094,
      "eval_loss": 0.2501136362552643,
      "eval_runtime": 7.3325,
      "eval_samples_per_second": 136.379,
      "eval_steps_per_second": 4.364,
      "step": 200
    },
    {
      "epoch": 1.01,
      "eval_accuracy": 0.9570000171661377,
      "eval_loss": 0.18152756989002228,
      "eval_runtime": 4.1659,
      "eval_samples_per_second": 240.043,
      "eval_steps_per_second": 7.681,
      "step": 300
    },
    {
      "epoch": 1.35,
      "eval_accuracy": 0.9649999737739563,
      "eval_loss": 0.1706397831439972,
      "eval_runtime": 4.1609,
      "eval_samples_per_second": 240.333,
      "eval_steps_per_second": 7.691,
      "step": 400
    },
    {
      "epoch": 1.69,
      "learning_rate": 8.738738738738739e-06,
      "loss": 0.0887,
      "step": 500
    },
    {
      "epoch": 1.69,
      "eval_accuracy": 0.9610000252723694,
      "eval_loss": 0.14937655627727509,
      "eval_runtime": 8.5296,
      "eval_samples_per_second": 117.239,
      "eval_steps_per_second": 3.752,
      "step": 500
    },
    {
      "epoch": 2.03,
      "eval_accuracy": 0.9660000205039978,
      "eval_loss": 0.14822959899902344,
      "eval_runtime": 7.4689,
      "eval_samples_per_second": 133.888,
      "eval_steps_per_second": 4.284,
      "step": 600
    },
    {
      "epoch": 2.36,
      "eval_accuracy": 0.9739999771118164,
      "eval_loss": 0.1305142641067505,
      "eval_runtime": 4.1624,
      "eval_samples_per_second": 240.248,
      "eval_steps_per_second": 7.688,
      "step": 700
    },
    {
      "epoch": 2.7,
      "eval_accuracy": 0.9639999866485596,
      "eval_loss": 0.14460867643356323,
      "eval_runtime": 4.1529,
      "eval_samples_per_second": 240.797,
      "eval_steps_per_second": 7.706,
      "step": 800
    },
    {
      "epoch": 3.0,
      "step": 888,
      "total_flos": 6921827938826112.0,
      "train_loss": 0.06248773110879434,
      "train_runtime": 2406.595,
      "train_samples_per_second": 11.79,
      "train_steps_per_second": 0.369
    }
  ],
  "max_steps": 888,
  "num_train_epochs": 3,
  "total_flos": 6921827938826112.0,
  "trial_name": null,
  "trial_params": null
}