{
  "best_metric": 0.5640622093599114,
  "best_model_checkpoint": "models_intermediate/intermediate_it5-large/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": 3.656,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "learning_rate": 3.8400000000000005e-05,
      "loss": 1.506,
      "step": 2000
    },
    {
      "epoch": 1.5,
      "learning_rate": 3.76e-05,
      "loss": 0.676,
      "step": 3000
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.680000000000001e-05,
      "loss": 0.4488,
      "step": 4000
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.6e-05,
      "loss": 0.3892,
      "step": 5000
    },
    {
      "epoch": 3.0,
      "learning_rate": 3.52e-05,
      "loss": 0.447,
      "step": 6000
    },
    {
      "epoch": 3.5,
      "learning_rate": 3.44e-05,
      "loss": 0.384,
      "step": 7000
    },
    {
      "epoch": 4.0,
      "learning_rate": 3.3600000000000004e-05,
      "loss": 0.346,
      "step": 8000
    },
    {
      "epoch": 4.5,
      "learning_rate": 3.28e-05,
      "loss": 0.3349,
      "step": 9000
    },
    {
      "epoch": 5.0,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 0.3415,
      "step": 10000
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.2377464771270752,
      "eval_p-value": 3.8666537188825695e-181,
      "eval_runtime": 246.1373,
      "eval_samples_per_second": 16.251,
      "eval_spearman": 0.4315122856249831,
      "eval_steps_per_second": 2.031,
      "step": 10000
    },
    {
      "epoch": 5.5,
      "learning_rate": 3.1200000000000006e-05,
      "loss": 0.316,
      "step": 11000
    },
    {
      "epoch": 6.0,
      "learning_rate": 3.0400000000000004e-05,
      "loss": 0.3282,
      "step": 12000
    },
    {
      "epoch": 6.5,
      "learning_rate": 2.96e-05,
      "loss": 0.2925,
      "step": 13000
    },
    {
      "epoch": 7.0,
      "learning_rate": 2.8800000000000002e-05,
      "loss": 0.2841,
      "step": 14000
    },
    {
      "epoch": 7.5,
      "learning_rate": 2.8e-05,
      "loss": 0.2719,
      "step": 15000
    },
    {
      "epoch": 8.0,
      "learning_rate": 2.7200000000000004e-05,
      "loss": 0.2694,
      "step": 16000
    },
    {
      "epoch": 8.5,
      "learning_rate": 2.6400000000000005e-05,
      "loss": 0.2592,
      "step": 17000
    },
    {
      "epoch": 9.0,
      "learning_rate": 2.5600000000000002e-05,
      "loss": 0.2563,
      "step": 18000
    },
    {
      "epoch": 9.5,
      "learning_rate": 2.4800000000000003e-05,
      "loss": 0.2599,
      "step": 19000
    },
    {
      "epoch": 10.0,
      "learning_rate": 2.4e-05,
      "loss": 0.2412,
      "step": 20000
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.19882595539093018,
      "eval_p-value": 2.0894786180148802e-236,
      "eval_runtime": 237.5854,
      "eval_samples_per_second": 16.836,
      "eval_spearman": 0.4861580881492778,
      "eval_steps_per_second": 2.105,
      "step": 20000
    },
    {
      "epoch": 10.5,
      "learning_rate": 2.32e-05,
      "loss": 0.2347,
      "step": 21000
    },
    {
      "epoch": 11.0,
      "learning_rate": 2.2400000000000002e-05,
      "loss": 0.2332,
      "step": 22000
    },
    {
      "epoch": 11.5,
      "learning_rate": 2.1600000000000003e-05,
      "loss": 0.2299,
      "step": 23000
    },
    {
      "epoch": 12.0,
      "learning_rate": 2.08e-05,
      "loss": 0.2258,
      "step": 24000
    },
    {
      "epoch": 12.5,
      "learning_rate": 2e-05,
      "loss": 0.2211,
      "step": 25000
    },
    {
      "epoch": 13.0,
      "learning_rate": 1.9200000000000003e-05,
      "loss": 0.223,
      "step": 26000
    },
    {
      "epoch": 13.5,
      "learning_rate": 1.8400000000000003e-05,
      "loss": 0.2201,
      "step": 27000
    },
    {
      "epoch": 14.0,
      "learning_rate": 1.76e-05,
      "loss": 0.2131,
      "step": 28000
    },
    {
      "epoch": 14.5,
      "learning_rate": 1.6800000000000002e-05,
      "loss": 0.2131,
      "step": 29000
    },
    {
      "epoch": 15.0,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.2153,
      "step": 30000
    },
    {
      "epoch": 15.0,
      "eval_loss": 0.18797849118709564,
      "eval_p-value": 5.983697306181229e-266,
      "eval_runtime": 248.0218,
      "eval_samples_per_second": 16.128,
      "eval_spearman": 0.511742524682024,
      "eval_steps_per_second": 2.016,
      "step": 30000
    },
    {
      "epoch": 15.5,
      "learning_rate": 1.5200000000000002e-05,
      "loss": 0.4026,
      "step": 31000
    },
    {
      "epoch": 16.0,
      "learning_rate": 1.4400000000000001e-05,
      "loss": 0.2148,
      "step": 32000
    },
    {
      "epoch": 16.5,
      "learning_rate": 1.3600000000000002e-05,
      "loss": 0.208,
      "step": 33000
    },
    {
      "epoch": 17.0,
      "learning_rate": 1.2800000000000001e-05,
      "loss": 0.2121,
      "step": 34000
    },
    {
      "epoch": 17.5,
      "learning_rate": 1.2e-05,
      "loss": 0.2064,
      "step": 35000
    },
    {
      "epoch": 18.0,
      "learning_rate": 1.1200000000000001e-05,
      "loss": 0.2075,
      "step": 36000
    },
    {
      "epoch": 18.5,
      "learning_rate": 1.04e-05,
      "loss": 0.2095,
      "step": 37000
    },
    {
      "epoch": 19.0,
      "learning_rate": 9.600000000000001e-06,
      "loss": 0.2058,
      "step": 38000
    },
    {
      "epoch": 19.5,
      "learning_rate": 8.8e-06,
      "loss": 0.2045,
      "step": 39000
    },
    {
      "epoch": 20.0,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.2054,
      "step": 40000
    },
    {
      "epoch": 20.0,
      "eval_loss": 0.1777072548866272,
      "eval_p-value": 8.897490187001762e-304,
      "eval_runtime": 252.1853,
      "eval_samples_per_second": 15.861,
      "eval_spearman": 0.541600893005368,
      "eval_steps_per_second": 1.983,
      "step": 40000
    },
    {
      "epoch": 20.5,
      "learning_rate": 7.2000000000000005e-06,
      "loss": 0.2044,
      "step": 41000
    },
    {
      "epoch": 21.0,
      "learning_rate": 6.4000000000000006e-06,
      "loss": 0.2003,
      "step": 42000
    },
    {
      "epoch": 21.5,
      "learning_rate": 5.600000000000001e-06,
      "loss": 0.1978,
      "step": 43000
    },
    {
      "epoch": 22.0,
      "learning_rate": 4.800000000000001e-06,
      "loss": 0.2044,
      "step": 44000
    },
    {
      "epoch": 22.5,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.1994,
      "step": 45000
    },
    {
      "epoch": 23.0,
      "learning_rate": 3.2000000000000003e-06,
      "loss": 0.1972,
      "step": 46000
    },
    {
      "epoch": 23.5,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 0.2022,
      "step": 47000
    },
    {
      "epoch": 24.0,
      "learning_rate": 1.6000000000000001e-06,
      "loss": 0.1955,
      "step": 48000
    },
    {
      "epoch": 24.5,
      "learning_rate": 8.000000000000001e-07,
      "loss": 0.1988,
      "step": 49000
    },
    {
      "epoch": 25.0,
      "learning_rate": 0.0,
      "loss": 0.1967,
      "step": 50000
    },
    {
      "epoch": 25.0,
      "eval_loss": 0.17539779841899872,
      "eval_p-value": 0.0,
      "eval_runtime": 247.6044,
      "eval_samples_per_second": 16.155,
      "eval_spearman": 0.5640622093599114,
      "eval_steps_per_second": 2.019,
      "step": 50000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 50000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 25,
  "save_steps": 10000,
  "total_flos": 4.609378615296e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}