trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 397,
  "global_step": 2384,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.17,
      "eval_gen_len": 81.6102,
      "eval_loss": 1.3396695852279663,
      "eval_rouge1": 52.6908,
      "eval_rouge2": 34.3367,
      "eval_rougeL": 43.9351,
      "eval_rougeLsum": 44.0153,
      "eval_runtime": 41.4692,
      "eval_samples_per_second": 1.423,
      "eval_steps_per_second": 0.723,
      "step": 397
    },
    {
      "epoch": 0.21,
      "grad_norm": 6.220447063446045,
      "learning_rate": 1.895763422818792e-05,
      "loss": 0.719,
      "step": 500
    },
    {
      "epoch": 0.33,
      "eval_gen_len": 84.2203,
      "eval_loss": 1.363059639930725,
      "eval_rouge1": 54.543,
      "eval_rouge2": 36.4199,
      "eval_rougeL": 45.8273,
      "eval_rougeLsum": 45.7925,
      "eval_runtime": 42.2281,
      "eval_samples_per_second": 1.397,
      "eval_steps_per_second": 0.71,
      "step": 794
    },
    {
      "epoch": 0.42,
      "grad_norm": 8.490495681762695,
      "learning_rate": 1.7908976510067115e-05,
      "loss": 0.7459,
      "step": 1000
    },
    {
      "epoch": 0.5,
      "eval_gen_len": 85.2373,
      "eval_loss": 1.3582559823989868,
      "eval_rouge1": 53.2262,
      "eval_rouge2": 34.8889,
      "eval_rougeL": 44.1043,
      "eval_rougeLsum": 44.0998,
      "eval_runtime": 42.6328,
      "eval_samples_per_second": 1.384,
      "eval_steps_per_second": 0.704,
      "step": 1191
    },
    {
      "epoch": 0.63,
      "grad_norm": 6.794929027557373,
      "learning_rate": 1.686031879194631e-05,
      "loss": 0.7154,
      "step": 1500
    },
    {
      "epoch": 0.67,
      "eval_gen_len": 83.7797,
      "eval_loss": 1.3886514902114868,
      "eval_rouge1": 54.9928,
      "eval_rouge2": 37.1125,
      "eval_rougeL": 46.4105,
      "eval_rougeLsum": 46.4044,
      "eval_runtime": 42.3464,
      "eval_samples_per_second": 1.393,
      "eval_steps_per_second": 0.708,
      "step": 1588
    },
    {
      "epoch": 0.83,
      "eval_gen_len": 85.8814,
      "eval_loss": 1.3405011892318726,
      "eval_rouge1": 52.5543,
      "eval_rouge2": 33.702,
      "eval_rougeL": 42.9428,
      "eval_rougeLsum": 43.0015,
      "eval_runtime": 43.1199,
      "eval_samples_per_second": 1.368,
      "eval_steps_per_second": 0.696,
      "step": 1985
    },
    {
      "epoch": 0.84,
      "grad_norm": 8.691970825195312,
      "learning_rate": 1.5811661073825504e-05,
      "loss": 0.7507,
      "step": 2000
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 81.7797,
      "eval_loss": 1.3399206399917603,
      "eval_rouge1": 52.4327,
      "eval_rouge2": 34.1158,
      "eval_rougeL": 43.2742,
      "eval_rougeLsum": 43.1693,
      "eval_runtime": 41.935,
      "eval_samples_per_second": 1.407,
      "eval_steps_per_second": 0.715,
      "step": 2382
    }
  ],
  "logging_steps": 500,
  "max_steps": 9536,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 1192,
  "total_flos": 5.17811143698432e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}