last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 2172,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_gen_len": 11.021739130434783,
      "eval_loss": 5.008267879486084,
      "eval_rouge1": 29.3556,
      "eval_rouge2": 23.1164,
      "eval_rougeL": 26.8076,
      "eval_rougeLsum": 26.8263,
      "eval_runtime": 173.025,
      "eval_samples_per_second": 1.861,
      "eval_steps_per_second": 0.237,
      "step": 362
    },
    {
      "epoch": 1.38,
      "learning_rate": 3e-05,
      "loss": 19.6995,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 10.335403726708075,
      "eval_loss": 0.8298134207725525,
      "eval_rouge1": 32.0346,
      "eval_rouge2": 24.9074,
      "eval_rougeL": 28.9355,
      "eval_rougeLsum": 29.0842,
      "eval_runtime": 150.6955,
      "eval_samples_per_second": 2.137,
      "eval_steps_per_second": 0.272,
      "step": 724
    },
    {
      "epoch": 2.76,
      "learning_rate": 2.5192307692307694e-05,
      "loss": 0.9076,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_gen_len": 10.863354037267081,
      "eval_loss": 0.7763211131095886,
      "eval_rouge1": 31.1555,
      "eval_rouge2": 23.6164,
      "eval_rougeL": 27.9825,
      "eval_rougeLsum": 28.1607,
      "eval_runtime": 162.2266,
      "eval_samples_per_second": 1.985,
      "eval_steps_per_second": 0.253,
      "step": 1086
    },
    {
      "epoch": 4.0,
      "eval_gen_len": 12.23913043478261,
      "eval_loss": 0.7397241592407227,
      "eval_rouge1": 28.3909,
      "eval_rouge2": 20.1283,
      "eval_rougeL": 24.9491,
      "eval_rougeLsum": 25.0633,
      "eval_runtime": 160.7559,
      "eval_samples_per_second": 2.003,
      "eval_steps_per_second": 0.255,
      "step": 1448
    },
    {
      "epoch": 4.14,
      "learning_rate": 2.0384615384615387e-05,
      "loss": 0.8095,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "eval_gen_len": 11.658385093167702,
      "eval_loss": 0.7186636328697205,
      "eval_rouge1": 30.3081,
      "eval_rouge2": 22.0426,
      "eval_rougeL": 26.8549,
      "eval_rougeLsum": 27.0402,
      "eval_runtime": 161.9975,
      "eval_samples_per_second": 1.988,
      "eval_steps_per_second": 0.253,
      "step": 1810
    },
    {
      "epoch": 5.52,
      "learning_rate": 1.557692307692308e-05,
      "loss": 0.7683,
      "step": 2000
    },
    {
      "epoch": 6.0,
      "eval_gen_len": 11.70807453416149,
      "eval_loss": 0.7056237459182739,
      "eval_rouge1": 30.3901,
      "eval_rouge2": 21.9556,
      "eval_rougeL": 26.8053,
      "eval_rougeLsum": 27.0162,
      "eval_runtime": 161.96,
      "eval_samples_per_second": 1.988,
      "eval_steps_per_second": 0.253,
      "step": 2172
    }
  ],
  "logging_steps": 500,
  "max_steps": 3620,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 1.164912412852224e+16,
  "trial_name": null,
  "trial_params": null
}