checkpoint-800/trainer_state.json

{
  "best_metric": 0.23951096832752228,
  "best_model_checkpoint": "./text2sql/codellama_instruct_spider_e10/checkpoint-500",
  "epoch": 2.956120092378753,
  "eval_steps": 100,
  "global_step": 800,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 3.08641975308642e-07,
      "loss": 1.3897,
      "step": 1
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.54320987654321e-05,
      "loss": 1.3449,
      "step": 50
    },
    {
      "epoch": 0.37,
      "learning_rate": 3.08641975308642e-05,
      "loss": 0.822,
      "step": 100
    },
    {
      "epoch": 0.37,
      "eval_loss": 0.5313276648521423,
      "eval_runtime": 34.4984,
      "eval_samples_per_second": 29.972,
      "eval_steps_per_second": 3.768,
      "step": 100
    },
    {
      "epoch": 0.55,
      "learning_rate": 4.62962962962963e-05,
      "loss": 0.4342,
      "step": 150
    },
    {
      "epoch": 0.74,
      "learning_rate": 4.9251379038613084e-05,
      "loss": 0.3014,
      "step": 200
    },
    {
      "epoch": 0.74,
      "eval_loss": 0.2763093411922455,
      "eval_runtime": 34.3952,
      "eval_samples_per_second": 30.062,
      "eval_steps_per_second": 3.78,
      "step": 200
    },
    {
      "epoch": 0.92,
      "learning_rate": 4.826635145784082e-05,
      "loss": 0.2401,
      "step": 250
    },
    {
      "epoch": 1.11,
      "learning_rate": 4.728132387706856e-05,
      "loss": 0.2091,
      "step": 300
    },
    {
      "epoch": 1.11,
      "eval_loss": 0.24685852229595184,
      "eval_runtime": 34.5214,
      "eval_samples_per_second": 29.952,
      "eval_steps_per_second": 3.766,
      "step": 300
    },
    {
      "epoch": 1.29,
      "learning_rate": 4.62962962962963e-05,
      "loss": 0.1837,
      "step": 350
    },
    {
      "epoch": 1.48,
      "learning_rate": 4.5311268715524035e-05,
      "loss": 0.1697,
      "step": 400
    },
    {
      "epoch": 1.48,
      "eval_loss": 0.2400880753993988,
      "eval_runtime": 34.3334,
      "eval_samples_per_second": 30.116,
      "eval_steps_per_second": 3.786,
      "step": 400
    },
    {
      "epoch": 1.66,
      "learning_rate": 4.432624113475177e-05,
      "loss": 0.1589,
      "step": 450
    },
    {
      "epoch": 1.85,
      "learning_rate": 4.334121355397952e-05,
      "loss": 0.1495,
      "step": 500
    },
    {
      "epoch": 1.85,
      "eval_loss": 0.23951096832752228,
      "eval_runtime": 34.3515,
      "eval_samples_per_second": 30.101,
      "eval_steps_per_second": 3.784,
      "step": 500
    },
    {
      "epoch": 2.03,
      "learning_rate": 4.235618597320725e-05,
      "loss": 0.1356,
      "step": 550
    },
    {
      "epoch": 2.22,
      "learning_rate": 4.1371158392434986e-05,
      "loss": 0.1256,
      "step": 600
    },
    {
      "epoch": 2.22,
      "eval_loss": 0.2525351345539093,
      "eval_runtime": 34.4715,
      "eval_samples_per_second": 29.996,
      "eval_steps_per_second": 3.771,
      "step": 600
    },
    {
      "epoch": 2.4,
      "learning_rate": 4.0386130811662727e-05,
      "loss": 0.1162,
      "step": 650
    },
    {
      "epoch": 2.59,
      "learning_rate": 3.940110323089047e-05,
      "loss": 0.1097,
      "step": 700
    },
    {
      "epoch": 2.59,
      "eval_loss": 0.2641294002532959,
      "eval_runtime": 34.4325,
      "eval_samples_per_second": 30.03,
      "eval_steps_per_second": 3.776,
      "step": 700
    },
    {
      "epoch": 2.77,
      "learning_rate": 3.84160756501182e-05,
      "loss": 0.1059,
      "step": 750
    },
    {
      "epoch": 2.96,
      "learning_rate": 3.743104806934594e-05,
      "loss": 0.1107,
      "step": 800
    },
    {
      "epoch": 2.96,
      "eval_loss": 0.2617126703262329,
      "eval_runtime": 34.3286,
      "eval_samples_per_second": 30.121,
      "eval_steps_per_second": 3.787,
      "step": 800
    }
  ],
  "logging_steps": 50,
  "max_steps": 2700,
  "num_train_epochs": 10,
  "save_steps": 100,
  "total_flos": 2.4444004623876096e+17,
  "trial_name": null,
  "trial_params": null
}