trainer_state.json

{
  "best_metric": 39.0802,
  "best_model_checkpoint": "bin/liputan6-base/checkpoint-252",
  "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": 6.927645683288574,
      "learning_rate": 0.0008,
      "loss": 1.6488,
      "step": 63
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 65.191,
      "eval_loss": 0.7318153381347656,
      "eval_rouge1": 34.224,
      "eval_rouge2": 24.5266,
      "eval_rougeL": 31.0318,
      "eval_rougeLsum": 32.8875,
      "eval_runtime": 255.5421,
      "eval_samples_per_second": 3.913,
      "eval_steps_per_second": 0.125,
      "step": 63
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.6224966049194336,
      "learning_rate": 0.0006,
      "loss": 0.6983,
      "step": 126
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 65.46,
      "eval_loss": 0.6432784795761108,
      "eval_rouge1": 37.3155,
      "eval_rouge2": 27.3019,
      "eval_rougeL": 33.9529,
      "eval_rougeLsum": 36.1013,
      "eval_runtime": 250.329,
      "eval_samples_per_second": 3.995,
      "eval_steps_per_second": 0.128,
      "step": 126
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.090550184249878,
      "learning_rate": 0.0004,
      "loss": 0.4226,
      "step": 189
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 59.969,
      "eval_loss": 0.5831208229064941,
      "eval_rouge1": 36.9679,
      "eval_rouge2": 26.3535,
      "eval_rougeL": 33.5956,
      "eval_rougeLsum": 35.7604,
      "eval_runtime": 241.9062,
      "eval_samples_per_second": 4.134,
      "eval_steps_per_second": 0.132,
      "step": 189
    },
    {
      "epoch": 4.0,
      "grad_norm": 4.525897026062012,
      "learning_rate": 0.0002,
      "loss": 0.242,
      "step": 252
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 55.301,
      "eval_loss": 0.5539225339889526,
      "eval_rouge1": 39.0802,
      "eval_rouge2": 28.4622,
      "eval_rougeL": 35.8085,
      "eval_rougeLsum": 37.8181,
      "eval_runtime": 216.2243,
      "eval_samples_per_second": 4.625,
      "eval_steps_per_second": 0.148,
      "step": 252
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.9606621861457825,
      "learning_rate": 0.0,
      "loss": 0.1248,
      "step": 315
    },
    {
      "epoch": 5.0,
      "eval_gen_len": 56.589,
      "eval_loss": 0.5169788599014282,
      "eval_rouge1": 38.108,
      "eval_rouge2": 27.5573,
      "eval_rougeL": 34.7198,
      "eval_rougeLsum": 36.6919,
      "eval_runtime": 211.3459,
      "eval_samples_per_second": 4.732,
      "eval_steps_per_second": 0.151,
      "step": 315
    },
    {
      "epoch": 5.0,
      "step": 315,
      "total_flos": 3423786762240000.0,
      "train_loss": 0.6272867081657288,
      "train_runtime": 1485.4068,
      "train_samples_per_second": 3.366,
      "train_steps_per_second": 0.212
    }
  ],
  "logging_steps": 500,
  "max_steps": 315,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 3423786762240000.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}