trainer_state.json

{
  "best_metric": 44.3826,
  "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": 1.0219109058380127,
      "learning_rate": 0.0008,
      "loss": 0.9013,
      "step": 63
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 46.887,
      "eval_loss": 0.36000651121139526,
      "eval_rouge1": 40.5674,
      "eval_rouge2": 32.5892,
      "eval_rougeL": 37.7471,
      "eval_rougeLsum": 39.1368,
      "eval_runtime": 220.7029,
      "eval_samples_per_second": 4.531,
      "eval_steps_per_second": 0.145,
      "step": 63
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.3100820779800415,
      "learning_rate": 0.0006,
      "loss": 0.4754,
      "step": 126
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 56.767,
      "eval_loss": 0.29582464694976807,
      "eval_rouge1": 43.3031,
      "eval_rouge2": 34.5149,
      "eval_rougeL": 39.7514,
      "eval_rougeLsum": 41.863,
      "eval_runtime": 188.9607,
      "eval_samples_per_second": 5.292,
      "eval_steps_per_second": 0.169,
      "step": 126
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.5092746019363403,
      "learning_rate": 0.0004,
      "loss": 0.3811,
      "step": 189
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 57.898,
      "eval_loss": 0.26292842626571655,
      "eval_rouge1": 43.4511,
      "eval_rouge2": 34.6775,
      "eval_rougeL": 39.9831,
      "eval_rougeLsum": 42.0606,
      "eval_runtime": 182.6393,
      "eval_samples_per_second": 5.475,
      "eval_steps_per_second": 0.175,
      "step": 189
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.0736172199249268,
      "learning_rate": 0.0002,
      "loss": 0.3317,
      "step": 252
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 57.815,
      "eval_loss": 0.2609642446041107,
      "eval_rouge1": 43.9259,
      "eval_rouge2": 35.2198,
      "eval_rougeL": 40.3143,
      "eval_rougeLsum": 42.5364,
      "eval_runtime": 186.4402,
      "eval_samples_per_second": 5.364,
      "eval_steps_per_second": 0.172,
      "step": 252
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.079635739326477,
      "learning_rate": 0.0,
      "loss": 0.299,
      "step": 315
    },
    {
      "epoch": 5.0,
      "eval_gen_len": 58.818,
      "eval_loss": 0.2554154396057129,
      "eval_rouge1": 44.3826,
      "eval_rouge2": 35.7034,
      "eval_rougeL": 40.7597,
      "eval_rougeLsum": 42.9985,
      "eval_runtime": 186.4288,
      "eval_samples_per_second": 5.364,
      "eval_steps_per_second": 0.172,
      "step": 315
    },
    {
      "epoch": 5.0,
      "step": 315,
      "total_flos": 3451266662400000.0,
      "train_loss": 0.477710209195576,
      "train_runtime": 1217.2314,
      "train_samples_per_second": 4.108,
      "train_steps_per_second": 0.259
    }
  ],
  "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
}