trainer_state.json

{
  "best_metric": 26.9743,
  "best_model_checkpoint": "bin/liputan6-seq_bn-rf16/checkpoint-315",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 315,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.7438846826553345,
      "learning_rate": 0.0008,
      "loss": 2.6241,
      "step": 63
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 33.387,
      "eval_loss": 2.753352403640747,
      "eval_rouge1": 23.3287,
      "eval_rouge2": 9.5988,
      "eval_rougeL": 20.1923,
      "eval_rougeLsum": 21.2916,
      "eval_runtime": 101.8941,
      "eval_samples_per_second": 9.814,
      "eval_steps_per_second": 0.314,
      "step": 63
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.6395297050476074,
      "learning_rate": 0.0006,
      "loss": 2.228,
      "step": 126
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 32.153,
      "eval_loss": 2.702471971511841,
      "eval_rouge1": 25.8033,
      "eval_rouge2": 10.8168,
      "eval_rougeL": 21.9451,
      "eval_rougeLsum": 23.4491,
      "eval_runtime": 112.7526,
      "eval_samples_per_second": 8.869,
      "eval_steps_per_second": 0.284,
      "step": 126
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.6557146310806274,
      "learning_rate": 0.0004,
      "loss": 2.0615,
      "step": 189
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 30.873,
      "eval_loss": 2.674882411956787,
      "eval_rouge1": 25.8887,
      "eval_rouge2": 10.7586,
      "eval_rougeL": 22.113,
      "eval_rougeLsum": 23.8997,
      "eval_runtime": 90.651,
      "eval_samples_per_second": 11.031,
      "eval_steps_per_second": 0.353,
      "step": 189
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.0358853340148926,
      "learning_rate": 0.0002,
      "loss": 1.9099,
      "step": 252
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 31.524,
      "eval_loss": 2.719660997390747,
      "eval_rouge1": 26.5565,
      "eval_rouge2": 11.2255,
      "eval_rougeL": 22.6026,
      "eval_rougeLsum": 24.5495,
      "eval_runtime": 95.5774,
      "eval_samples_per_second": 10.463,
      "eval_steps_per_second": 0.335,
      "step": 252
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.7301920652389526,
      "learning_rate": 0.0,
      "loss": 1.8007,
      "step": 315
    },
    {
      "epoch": 5.0,
      "eval_gen_len": 33.854,
      "eval_loss": 2.7479429244995117,
      "eval_rouge1": 26.9743,
      "eval_rouge2": 11.4843,
      "eval_rougeL": 22.9863,
      "eval_rougeLsum": 24.9284,
      "eval_runtime": 101.7272,
      "eval_samples_per_second": 9.83,
      "eval_steps_per_second": 0.315,
      "step": 315
    },
    {
      "epoch": 5.0,
      "step": 315,
      "total_flos": 3451266662400000.0,
      "train_loss": 2.1248473152281746,
      "train_runtime": 753.9081,
      "train_samples_per_second": 6.632,
      "train_steps_per_second": 0.418
    }
  ],
  "logging_steps": 500,
  "max_steps": 315,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 3451266662400000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}