{
  "best_metric": 0.7703425408958412,
  "best_model_checkpoint": "models_intermediate/intermediate_it5-base/checkpoint-50000",
  "epoch": 25.0,
  "eval_steps": 10000,
  "global_step": 50000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5,
      "learning_rate": 3.9200000000000004e-05,
      "loss": 3.7449,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "learning_rate": 3.8400000000000005e-05,
      "loss": 0.4206,
      "step": 2000
    },
    {
      "epoch": 1.5,
      "learning_rate": 3.76e-05,
      "loss": 0.3315,
      "step": 3000
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.680000000000001e-05,
      "loss": 0.2858,
      "step": 4000
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.6e-05,
      "loss": 0.2659,
      "step": 5000
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.52e-05,
      "loss": 0.2467,
      "step": 6000
    },
    {
      "epoch": 3.5,
      "learning_rate": 3.44e-05,
      "loss": 0.231,
      "step": 7000
    },
    {
      "epoch": 4.0,
      "learning_rate": 3.3600000000000004e-05,
      "loss": 0.233,
      "step": 8000
    },
    {
      "epoch": 4.5,
      "learning_rate": 3.28e-05,
      "loss": 0.2216,
      "step": 9000
    },
    {
      "epoch": 5.0,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 0.2117,
      "step": 10000
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.18562966585159302,
      "eval_p-value": 0.0,
      "eval_runtime": 91.1994,
      "eval_samples_per_second": 43.86,
      "eval_spearman": 0.5467240872218808,
      "eval_steps_per_second": 5.482,
      "step": 10000
    },
    {
      "epoch": 5.5,
      "learning_rate": 3.1200000000000006e-05,
      "loss": 0.2064,
      "step": 11000
    },
    {
      "epoch": 6.0,
      "learning_rate": 3.0400000000000004e-05,
      "loss": 0.2039,
      "step": 12000
    },
    {
      "epoch": 6.5,
      "learning_rate": 2.96e-05,
      "loss": 0.1986,
      "step": 13000
    },
    {
      "epoch": 7.0,
      "learning_rate": 2.8800000000000002e-05,
      "loss": 0.1935,
      "step": 14000
    },
    {
      "epoch": 7.5,
      "learning_rate": 2.8e-05,
      "loss": 0.1918,
      "step": 15000
    },
    {
      "epoch": 8.0,
      "learning_rate": 2.7200000000000004e-05,
      "loss": 0.1848,
      "step": 16000
    },
    {
      "epoch": 8.5,
      "learning_rate": 2.6400000000000005e-05,
      "loss": 0.1854,
      "step": 17000
    },
    {
      "epoch": 9.0,
      "learning_rate": 2.5600000000000002e-05,
      "loss": 0.1826,
      "step": 18000
    },
    {
      "epoch": 9.5,
      "learning_rate": 2.4800000000000003e-05,
      "loss": 0.1725,
      "step": 19000
    },
    {
      "epoch": 10.0,
      "learning_rate": 2.4e-05,
      "loss": 0.1817,
      "step": 20000
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.16969460248947144,
      "eval_p-value": 0.0,
      "eval_runtime": 104.6636,
      "eval_samples_per_second": 38.218,
      "eval_spearman": 0.6838117625451964,
      "eval_steps_per_second": 4.777,
      "step": 20000
    },
    {
      "epoch": 10.5,
      "learning_rate": 2.32e-05,
      "loss": 0.1685,
      "step": 21000
    },
    {
      "epoch": 11.0,
      "learning_rate": 2.2400000000000002e-05,
      "loss": 0.1759,
      "step": 22000
    },
    {
      "epoch": 11.5,
      "learning_rate": 2.1600000000000003e-05,
      "loss": 0.17,
      "step": 23000
    },
    {
      "epoch": 12.0,
      "learning_rate": 2.08e-05,
      "loss": 0.1672,
      "step": 24000
    },
    {
      "epoch": 12.5,
      "learning_rate": 2e-05,
      "loss": 0.1606,
      "step": 25000
    },
    {
      "epoch": 13.0,
      "learning_rate": 1.9200000000000003e-05,
      "loss": 0.1669,
      "step": 26000
    },
    {
      "epoch": 13.5,
      "learning_rate": 1.8400000000000003e-05,
      "loss": 0.1615,
      "step": 27000
    },
    {
      "epoch": 14.0,
      "learning_rate": 1.76e-05,
      "loss": 0.1608,
      "step": 28000
    },
    {
      "epoch": 14.5,
      "learning_rate": 1.6800000000000002e-05,
      "loss": 0.1615,
      "step": 29000
    },
    {
      "epoch": 15.0,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.154,
      "step": 30000
    },
    {
      "epoch": 15.0,
      "eval_loss": 0.15203040838241577,
      "eval_p-value": 0.0,
      "eval_runtime": 105.6332,
      "eval_samples_per_second": 37.867,
      "eval_spearman": 0.7306849120858863,
      "eval_steps_per_second": 4.733,
      "step": 30000
    },
    {
      "epoch": 15.5,
      "learning_rate": 1.5200000000000002e-05,
      "loss": 0.1533,
      "step": 31000
    },
    {
      "epoch": 16.0,
      "learning_rate": 1.4400000000000001e-05,
      "loss": 0.1561,
      "step": 32000
    },
    {
      "epoch": 16.5,
      "learning_rate": 1.3600000000000002e-05,
      "loss": 0.1534,
      "step": 33000
    },
    {
      "epoch": 17.0,
      "learning_rate": 1.2800000000000001e-05,
      "loss": 0.1522,
      "step": 34000
    },
    {
      "epoch": 17.5,
      "learning_rate": 1.2e-05,
      "loss": 0.1492,
      "step": 35000
    },
    {
      "epoch": 18.0,
      "learning_rate": 1.1200000000000001e-05,
      "loss": 0.1516,
      "step": 36000
    },
    {
      "epoch": 18.5,
      "learning_rate": 1.04e-05,
      "loss": 0.1455,
      "step": 37000
    },
    {
      "epoch": 19.0,
      "learning_rate": 9.600000000000001e-06,
      "loss": 0.151,
      "step": 38000
    },
    {
      "epoch": 19.5,
      "learning_rate": 8.8e-06,
      "loss": 0.1454,
      "step": 39000
    },
    {
      "epoch": 20.0,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.1485,
      "step": 40000
    },
    {
      "epoch": 20.0,
      "eval_loss": 0.14644727110862732,
      "eval_p-value": 0.0,
      "eval_runtime": 209.1893,
      "eval_samples_per_second": 19.121,
      "eval_spearman": 0.7611950719272746,
      "eval_steps_per_second": 2.39,
      "step": 40000
    },
    {
      "epoch": 20.5,
      "learning_rate": 7.2000000000000005e-06,
      "loss": 0.1462,
      "step": 41000
    },
    {
      "epoch": 21.0,
      "learning_rate": 6.4000000000000006e-06,
      "loss": 0.1462,
      "step": 42000
    },
    {
      "epoch": 21.5,
      "learning_rate": 5.600000000000001e-06,
      "loss": 0.1406,
      "step": 43000
    },
    {
      "epoch": 22.0,
      "learning_rate": 4.800000000000001e-06,
      "loss": 0.1467,
      "step": 44000
    },
    {
      "epoch": 22.5,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.1428,
      "step": 45000
    },
    {
      "epoch": 23.0,
      "learning_rate": 3.2000000000000003e-06,
      "loss": 0.1436,
      "step": 46000
    },
    {
      "epoch": 23.5,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 0.1454,
      "step": 47000
    },
    {
      "epoch": 24.0,
      "learning_rate": 1.6000000000000001e-06,
      "loss": 0.1409,
      "step": 48000
    },
    {
      "epoch": 24.5,
      "learning_rate": 8.000000000000001e-07,
      "loss": 0.1399,
      "step": 49000
    },
    {
      "epoch": 25.0,
      "learning_rate": 0.0,
      "loss": 0.1435,
      "step": 50000
    },
    {
      "epoch": 25.0,
      "eval_loss": 0.14494818449020386,
      "eval_p-value": 0.0,
      "eval_runtime": 178.9079,
      "eval_samples_per_second": 22.358,
      "eval_spearman": 0.7703425408958412,
      "eval_steps_per_second": 2.795,
      "step": 50000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 50000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 25,
  "save_steps": 10000,
  "total_flos": 1.369396740096e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}