{
  "best_metric": 0.8445149587473081,
  "best_model_checkpoint": "models_intermediate/intermediate_t5-base_25/checkpoint-55000",
  "epoch": 27.5,
  "eval_steps": 11000,
  "global_step": 55000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5,
      "learning_rate": 3.927272727272728e-05,
      "loss": 0.6916,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "learning_rate": 3.854545454545455e-05,
      "loss": 0.366,
      "step": 2000
    },
    {
      "epoch": 1.5,
      "learning_rate": 3.7818181818181824e-05,
      "loss": 0.3026,
      "step": 3000
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.709090909090909e-05,
      "loss": 0.249,
      "step": 4000
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.6363636363636364e-05,
      "loss": 0.2207,
      "step": 5000
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.563636363636364e-05,
      "loss": 0.2124,
      "step": 6000
    },
    {
      "epoch": 3.5,
      "learning_rate": 3.490909090909091e-05,
      "loss": 0.1911,
      "step": 7000
    },
    {
      "epoch": 4.0,
      "learning_rate": 3.4181818181818185e-05,
      "loss": 0.1819,
      "step": 8000
    },
    {
      "epoch": 4.5,
      "learning_rate": 3.345454545454546e-05,
      "loss": 0.1666,
      "step": 9000
    },
    {
      "epoch": 5.0,
      "learning_rate": 3.272727272727273e-05,
      "loss": 0.164,
      "step": 10000
    },
    {
      "epoch": 5.5,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 0.1497,
      "step": 11000
    },
    {
      "epoch": 5.5,
      "eval_loss": 0.14336544275283813,
      "eval_p-value": 0.0,
      "eval_runtime": 137.1062,
      "eval_samples_per_second": 29.174,
      "eval_spearman": 0.6969240735042003,
      "eval_steps_per_second": 3.647,
      "step": 11000
    },
    {
      "epoch": 6.0,
      "learning_rate": 3.127272727272728e-05,
      "loss": 0.1481,
      "step": 12000
    },
    {
      "epoch": 6.5,
      "learning_rate": 3.054545454545455e-05,
      "loss": 0.1341,
      "step": 13000
    },
    {
      "epoch": 7.0,
      "learning_rate": 2.9818181818181823e-05,
      "loss": 0.1336,
      "step": 14000
    },
    {
      "epoch": 7.5,
      "learning_rate": 2.9090909090909093e-05,
      "loss": 0.1217,
      "step": 15000
    },
    {
      "epoch": 8.0,
      "learning_rate": 2.8363636363636366e-05,
      "loss": 0.1236,
      "step": 16000
    },
    {
      "epoch": 8.5,
      "learning_rate": 2.763636363636364e-05,
      "loss": 0.1156,
      "step": 17000
    },
    {
      "epoch": 9.0,
      "learning_rate": 2.690909090909091e-05,
      "loss": 0.1094,
      "step": 18000
    },
    {
      "epoch": 9.5,
      "learning_rate": 2.6181818181818183e-05,
      "loss": 0.1034,
      "step": 19000
    },
    {
      "epoch": 10.0,
      "learning_rate": 2.5454545454545457e-05,
      "loss": 0.1044,
      "step": 20000
    },
    {
      "epoch": 10.5,
      "learning_rate": 2.4727272727272727e-05,
      "loss": 0.0915,
      "step": 21000
    },
    {
      "epoch": 11.0,
      "learning_rate": 2.4e-05,
      "loss": 0.098,
      "step": 22000
    },
    {
      "epoch": 11.0,
      "eval_loss": 0.12715213000774384,
      "eval_p-value": 0.0,
      "eval_runtime": 146.4466,
      "eval_samples_per_second": 27.314,
      "eval_spearman": 0.8180091381714769,
      "eval_steps_per_second": 3.414,
      "step": 22000
    },
    {
      "epoch": 11.5,
      "learning_rate": 2.3272727272727274e-05,
      "loss": 0.0899,
      "step": 23000
    },
    {
      "epoch": 12.0,
      "learning_rate": 2.2545454545454544e-05,
      "loss": 0.0844,
      "step": 24000
    },
    {
      "epoch": 12.5,
      "learning_rate": 2.1818181818181818e-05,
      "loss": 0.082,
      "step": 25000
    },
    {
      "epoch": 13.0,
      "learning_rate": 2.109090909090909e-05,
      "loss": 0.0796,
      "step": 26000
    },
    {
      "epoch": 13.5,
      "learning_rate": 2.0363636363636365e-05,
      "loss": 0.075,
      "step": 27000
    },
    {
      "epoch": 14.0,
      "learning_rate": 1.963636363636364e-05,
      "loss": 0.0751,
      "step": 28000
    },
    {
      "epoch": 14.5,
      "learning_rate": 1.8909090909090912e-05,
      "loss": 0.0692,
      "step": 29000
    },
    {
      "epoch": 15.0,
      "learning_rate": 1.8181818181818182e-05,
      "loss": 0.0685,
      "step": 30000
    },
    {
      "epoch": 15.5,
      "learning_rate": 1.7454545454545456e-05,
      "loss": 0.0664,
      "step": 31000
    },
    {
      "epoch": 16.0,
      "learning_rate": 1.672727272727273e-05,
      "loss": 0.0624,
      "step": 32000
    },
    {
      "epoch": 16.5,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.0595,
      "step": 33000
    },
    {
      "epoch": 16.5,
      "eval_loss": 0.1381928026676178,
      "eval_p-value": 0.0,
      "eval_runtime": 72.1726,
      "eval_samples_per_second": 55.423,
      "eval_spearman": 0.8359601637273105,
      "eval_steps_per_second": 6.928,
      "step": 33000
    },
    {
      "epoch": 17.0,
      "learning_rate": 1.5272727272727276e-05,
      "loss": 0.057,
      "step": 34000
    },
    {
      "epoch": 17.5,
      "learning_rate": 1.4545454545454546e-05,
      "loss": 0.0567,
      "step": 35000
    },
    {
      "epoch": 18.0,
      "learning_rate": 1.381818181818182e-05,
      "loss": 0.0546,
      "step": 36000
    },
    {
      "epoch": 18.5,
      "learning_rate": 1.3090909090909092e-05,
      "loss": 0.0513,
      "step": 37000
    },
    {
      "epoch": 19.0,
      "learning_rate": 1.2363636363636364e-05,
      "loss": 0.053,
      "step": 38000
    },
    {
      "epoch": 19.5,
      "learning_rate": 1.1636363636363637e-05,
      "loss": 0.0495,
      "step": 39000
    },
    {
      "epoch": 20.0,
      "learning_rate": 1.0909090909090909e-05,
      "loss": 0.0484,
      "step": 40000
    },
    {
      "epoch": 20.5,
      "learning_rate": 1.0181818181818182e-05,
      "loss": 0.0435,
      "step": 41000
    },
    {
      "epoch": 21.0,
      "learning_rate": 9.454545454545456e-06,
      "loss": 0.0488,
      "step": 42000
    },
    {
      "epoch": 21.5,
      "learning_rate": 8.727272727272728e-06,
      "loss": 0.0412,
      "step": 43000
    },
    {
      "epoch": 22.0,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.0461,
      "step": 44000
    },
    {
      "epoch": 22.0,
      "eval_loss": 0.1516806036233902,
      "eval_p-value": 0.0,
      "eval_runtime": 72.484,
      "eval_samples_per_second": 55.185,
      "eval_spearman": 0.8438657018617839,
      "eval_steps_per_second": 6.898,
      "step": 44000
    },
    {
      "epoch": 22.5,
      "learning_rate": 7.272727272727273e-06,
      "loss": 0.0414,
      "step": 45000
    },
    {
      "epoch": 23.0,
      "learning_rate": 6.545454545454546e-06,
      "loss": 0.0423,
      "step": 46000
    },
    {
      "epoch": 23.5,
      "learning_rate": 5.8181818181818185e-06,
      "loss": 0.0403,
      "step": 47000
    },
    {
      "epoch": 24.0,
      "learning_rate": 5.090909090909091e-06,
      "loss": 0.0404,
      "step": 48000
    },
    {
      "epoch": 24.5,
      "learning_rate": 4.363636363636364e-06,
      "loss": 0.0374,
      "step": 49000
    },
    {
      "epoch": 25.0,
      "learning_rate": 3.6363636363636366e-06,
      "loss": 0.0405,
      "step": 50000
    },
    {
      "epoch": 25.5,
      "learning_rate": 2.9090909090909093e-06,
      "loss": 0.0382,
      "step": 51000
    },
    {
      "epoch": 26.0,
      "learning_rate": 2.181818181818182e-06,
      "loss": 0.0364,
      "step": 52000
    },
    {
      "epoch": 26.5,
      "learning_rate": 1.4545454545454546e-06,
      "loss": 0.0367,
      "step": 53000
    },
    {
      "epoch": 27.0,
      "learning_rate": 7.272727272727273e-07,
      "loss": 0.036,
      "step": 54000
    },
    {
      "epoch": 27.5,
      "learning_rate": 0.0,
      "loss": 0.0355,
      "step": 55000
    },
    {
      "epoch": 27.5,
      "eval_loss": 0.1589333862066269,
      "eval_p-value": 0.0,
      "eval_runtime": 72.5243,
      "eval_samples_per_second": 55.154,
      "eval_spearman": 0.8445149587473081,
      "eval_steps_per_second": 6.894,
      "step": 55000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 55000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 28,
  "save_steps": 11000,
  "total_flos": 1.3396899468423168e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}