trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "global_step": 20455,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 2.500229112570643e-05,
      "loss": 0.2994,
      "step": 4091
    },
    {
      "epoch": 1.0,
      "eval_HasAns_exact": 50.77597840755735,
      "eval_HasAns_f1": 59.77609298252823,
      "eval_HasAns_total": 5928,
      "eval_NoAns_exact": 78.60386879730866,
      "eval_NoAns_f1": 78.60386879730866,
      "eval_NoAns_total": 5945,
      "eval_best_exact": 64.7098458687779,
      "eval_best_exact_thresh": 0.0,
      "eval_best_f1": 69.20345988380656,
      "eval_best_f1_thresh": 0.0,
      "eval_exact": 64.7098458687779,
      "eval_f1": 69.20345988380627,
      "eval_runtime": 69.4653,
      "eval_samples_per_second": 172.748,
      "eval_steps_per_second": 5.398,
      "eval_total": 11873,
      "step": 4091
    },
    {
      "epoch": 2.0,
      "learning_rate": 1.875515503283947e-05,
      "loss": 0.0681,
      "step": 8182
    },
    {
      "epoch": 2.0,
      "eval_HasAns_exact": 54.97638326585695,
      "eval_HasAns_f1": 66.01412319490598,
      "eval_HasAns_total": 5928,
      "eval_NoAns_exact": 68.04037005887301,
      "eval_NoAns_f1": 68.04037005887301,
      "eval_NoAns_total": 5945,
      "eval_best_exact": 61.517729301777145,
      "eval_best_exact_thresh": 0.0,
      "eval_best_f1": 67.02869723737984,
      "eval_best_f1_thresh": 0.0,
      "eval_exact": 61.517729301777145,
      "eval_f1": 67.02869723737946,
      "eval_runtime": 69.813,
      "eval_samples_per_second": 171.888,
      "eval_steps_per_second": 5.371,
      "eval_total": 11873,
      "step": 8182
    },
    {
      "epoch": 3.0,
      "learning_rate": 1.2509546357110128e-05,
      "loss": 0.0373,
      "step": 12273
    },
    {
      "epoch": 3.0,
      "eval_HasAns_exact": 52.66531713900135,
      "eval_HasAns_f1": 63.1366903466774,
      "eval_HasAns_total": 5928,
      "eval_NoAns_exact": 74.23044575273339,
      "eval_NoAns_f1": 74.23044575273339,
      "eval_NoAns_total": 5945,
      "eval_best_exact": 63.46332013812853,
      "eval_best_exact_thresh": 0.0,
      "eval_best_f1": 68.6915101806715,
      "eval_best_f1_thresh": 0.0,
      "eval_exact": 63.46332013812853,
      "eval_f1": 68.6915101806711,
      "eval_runtime": 69.5169,
      "eval_samples_per_second": 172.62,
      "eval_steps_per_second": 5.394,
      "eval_total": 11873,
      "step": 12273
    },
    {
      "epoch": 4.0,
      "learning_rate": 6.262410264243165e-06,
      "loss": 0.0144,
      "step": 16364
    },
    {
      "epoch": 4.0,
      "eval_HasAns_exact": 53.07017543859649,
      "eval_HasAns_f1": 63.174613714318006,
      "eval_HasAns_total": 5928,
      "eval_NoAns_exact": 74.88645920941968,
      "eval_NoAns_f1": 74.88645920941968,
      "eval_NoAns_total": 5945,
      "eval_best_exact": 63.99393582076981,
      "eval_best_exact_thresh": 0.0,
      "eval_best_f1": 69.03892108974036,
      "eval_best_f1_thresh": 0.0,
      "eval_exact": 63.99393582076981,
      "eval_f1": 69.03892108973986,
      "eval_runtime": 69.3973,
      "eval_samples_per_second": 172.917,
      "eval_steps_per_second": 5.404,
      "eval_total": 11873,
      "step": 16364
    },
    {
      "epoch": 5.0,
      "learning_rate": 1.3746754238582558e-08,
      "loss": 0.0043,
      "step": 20455
    },
    {
      "epoch": 5.0,
      "eval_HasAns_exact": 53.54251012145749,
      "eval_HasAns_f1": 63.501893668646716,
      "eval_HasAns_total": 5928,
      "eval_NoAns_exact": 75.10513036164845,
      "eval_NoAns_f1": 75.10513036164845,
      "eval_NoAns_total": 5945,
      "eval_best_exact": 64.3392571380443,
      "eval_best_exact_thresh": 0.0,
      "eval_best_f1": 69.31181888888628,
      "eval_best_f1_thresh": 0.0,
      "eval_exact": 64.3392571380443,
      "eval_f1": 69.31181888888594,
      "eval_runtime": 69.762,
      "eval_samples_per_second": 172.014,
      "eval_steps_per_second": 5.375,
      "eval_total": 11873,
      "step": 20455
    },
    {
      "epoch": 5.0,
      "step": 20455,
      "total_flos": 1.445362831448064e+16,
      "train_loss": 0.08469501011414156,
      "train_runtime": 9458.5775,
      "train_samples_per_second": 69.195,
      "train_steps_per_second": 2.163
    }
  ],
  "max_steps": 20455,
  "num_train_epochs": 5,
  "total_flos": 1.445362831448064e+16,
  "trial_name": null,
  "trial_params": null
}