trainer_state.json

{
  "best_metric": 3.896470308303833,
  "best_model_checkpoint": "tam_test_output/checkpoint-24",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 24,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_gen_len": 19.0,
      "eval_loss": 4.7975006103515625,
      "eval_rouge1": 0.1363,
      "eval_rouge2": 0.047,
      "eval_rougeL": 0.114,
      "eval_rougeLsum": 0.1145,
      "eval_runtime": 0.7346,
      "eval_samples_per_second": 66.707,
      "eval_steps_per_second": 1.361,
      "step": 3
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 19.0,
      "eval_loss": 4.614617824554443,
      "eval_rouge1": 0.1365,
      "eval_rouge2": 0.0484,
      "eval_rougeL": 0.1143,
      "eval_rougeLsum": 0.115,
      "eval_runtime": 0.729,
      "eval_samples_per_second": 67.217,
      "eval_steps_per_second": 1.372,
      "step": 6
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 19.0,
      "eval_loss": 4.439556121826172,
      "eval_rouge1": 0.1375,
      "eval_rouge2": 0.049,
      "eval_rougeL": 0.1143,
      "eval_rougeLsum": 0.1149,
      "eval_runtime": 0.7323,
      "eval_samples_per_second": 66.914,
      "eval_steps_per_second": 1.366,
      "step": 9
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 19.0,
      "eval_loss": 4.277223110198975,
      "eval_rouge1": 0.1381,
      "eval_rouge2": 0.0497,
      "eval_rougeL": 0.1144,
      "eval_rougeLsum": 0.115,
      "eval_runtime": 0.7261,
      "eval_samples_per_second": 67.482,
      "eval_steps_per_second": 1.377,
      "step": 12
    },
    {
      "epoch": 5.0,
      "eval_gen_len": 19.0,
      "eval_loss": 4.136841297149658,
      "eval_rouge1": 0.1389,
      "eval_rouge2": 0.0505,
      "eval_rougeL": 0.1155,
      "eval_rougeLsum": 0.1161,
      "eval_runtime": 0.7477,
      "eval_samples_per_second": 65.538,
      "eval_steps_per_second": 1.338,
      "step": 15
    },
    {
      "epoch": 6.0,
      "eval_gen_len": 19.0,
      "eval_loss": 4.028151988983154,
      "eval_rouge1": 0.1397,
      "eval_rouge2": 0.0509,
      "eval_rougeL": 0.1159,
      "eval_rougeLsum": 0.1166,
      "eval_runtime": 0.7264,
      "eval_samples_per_second": 67.46,
      "eval_steps_per_second": 1.377,
      "step": 18
    },
    {
      "epoch": 7.0,
      "eval_gen_len": 19.0,
      "eval_loss": 3.948868989944458,
      "eval_rouge1": 0.1391,
      "eval_rouge2": 0.0509,
      "eval_rougeL": 0.1156,
      "eval_rougeLsum": 0.1161,
      "eval_runtime": 0.7261,
      "eval_samples_per_second": 67.481,
      "eval_steps_per_second": 1.377,
      "step": 21
    },
    {
      "epoch": 8.0,
      "eval_gen_len": 19.0,
      "eval_loss": 3.896470308303833,
      "eval_rouge1": 0.1391,
      "eval_rouge2": 0.0509,
      "eval_rougeL": 0.1156,
      "eval_rougeLsum": 0.1161,
      "eval_runtime": 0.744,
      "eval_samples_per_second": 65.864,
      "eval_steps_per_second": 1.344,
      "step": 24
    }
  ],
  "logging_steps": 500,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 426597358239744.0,
  "train_batch_size": 96,
  "trial_name": null,
  "trial_params": null
}