trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 370,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8234661221504211,
      "eval_f1": "[0.91702786 0.33333333 0.         0.10837438]",
      "eval_loss": 0.48905953764915466,
      "eval_runtime": 8.4871,
      "eval_samples_per_second": 218.92,
      "eval_steps_per_second": 3.535,
      "step": 37
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8320775032043457,
      "eval_f1": "[0.93139159 0.48466258 0.         0.22857143]",
      "eval_loss": 0.47622746229171753,
      "eval_runtime": 8.4411,
      "eval_samples_per_second": 220.115,
      "eval_steps_per_second": 3.554,
      "step": 74
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.820775032043457,
      "eval_f1": "[0.92995725 0.44887781 0.         0.19266055]",
      "eval_loss": 0.5084477066993713,
      "eval_runtime": 8.3479,
      "eval_samples_per_second": 222.571,
      "eval_steps_per_second": 3.594,
      "step": 111
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8105489611625671,
      "eval_f1": "[0.92421691 0.44444444 0.06557377 0.30769231]",
      "eval_loss": 0.5519338846206665,
      "eval_runtime": 8.1137,
      "eval_samples_per_second": 228.996,
      "eval_steps_per_second": 3.697,
      "step": 148
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8293864130973816,
      "eval_f1": "[0.93531353 0.52336449 0.09345794 0.27131783]",
      "eval_loss": 0.5805315971374512,
      "eval_runtime": 8.8001,
      "eval_samples_per_second": 211.135,
      "eval_steps_per_second": 3.409,
      "step": 185
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7954789996147156,
      "eval_f1": "[0.91344509 0.55305466 0.15463918 0.29166667]",
      "eval_loss": 0.6778322458267212,
      "eval_runtime": 7.8496,
      "eval_samples_per_second": 236.701,
      "eval_steps_per_second": 3.822,
      "step": 222
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8213132619857788,
      "eval_f1": "[0.93298292 0.51383399 0.10191083 0.2519084 ]",
      "eval_loss": 0.6407278776168823,
      "eval_runtime": 8.767,
      "eval_samples_per_second": 211.931,
      "eval_steps_per_second": 3.422,
      "step": 259
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.827233612537384,
      "eval_f1": "[0.9326288  0.55052265 0.18181818 0.26271186]",
      "eval_loss": 0.6639025807380676,
      "eval_runtime": 8.0873,
      "eval_samples_per_second": 229.742,
      "eval_steps_per_second": 3.709,
      "step": 296
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8191604018211365,
      "eval_f1": "[0.93071286 0.55830389 0.11042945 0.2761194 ]",
      "eval_loss": 0.6863259077072144,
      "eval_runtime": 7.9866,
      "eval_samples_per_second": 232.64,
      "eval_steps_per_second": 3.756,
      "step": 333
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.81431645154953,
      "eval_f1": "[0.92816572 0.56028369 0.1        0.2633452 ]",
      "eval_loss": 0.6886363625526428,
      "eval_runtime": 7.7254,
      "eval_samples_per_second": 240.504,
      "eval_steps_per_second": 3.883,
      "step": 370
    },
    {
      "epoch": 10.0,
      "step": 370,
      "total_flos": 1537911734661120.0,
      "train_loss": 0.21409231649862753,
      "train_runtime": 515.9059,
      "train_samples_per_second": 45.318,
      "train_steps_per_second": 0.717
    }
  ],
  "logging_steps": 500,
  "max_steps": 370,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 1537911734661120.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}