{
  "best_metric": 0.5610724925521351,
  "best_model_checkpoint": "./experiment/train_0519_cosql_order_relation_add_coref/checkpoint-576",
  "epoch": 71.9922480620155,
  "global_step": 576,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12,
      "learning_rate": 0.0001,
      "loss": 3.711,
      "step": 1
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0001,
      "loss": 2.4467,
      "step": 8
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.0001,
      "loss": 0.7539,
      "step": 16
    },
    {
      "epoch": 2.99,
      "learning_rate": 0.0001,
      "loss": 0.405,
      "step": 24
    },
    {
      "epoch": 3.99,
      "learning_rate": 0.0001,
      "loss": 0.3139,
      "step": 32
    },
    {
      "epoch": 4.99,
      "learning_rate": 0.0001,
      "loss": 0.2368,
      "step": 40
    },
    {
      "epoch": 5.99,
      "learning_rate": 0.0001,
      "loss": 0.2018,
      "step": 48
    },
    {
      "epoch": 6.99,
      "learning_rate": 0.0001,
      "loss": 0.1558,
      "step": 56
    },
    {
      "epoch": 7.99,
      "learning_rate": 0.0001,
      "loss": 0.149,
      "step": 64
    },
    {
      "epoch": 7.99,
      "eval_exact_match": 0.4240317775571003,
      "eval_exec": 0.5024826216484608,
      "eval_loss": 0.2136896252632141,
      "eval_runtime": 782.5228,
      "eval_samples_per_second": 1.661,
      "step": 64
    },
    {
      "epoch": 8.99,
      "learning_rate": 0.0001,
      "loss": 0.1223,
      "step": 72
    },
    {
      "epoch": 9.99,
      "learning_rate": 0.0001,
      "loss": 0.1069,
      "step": 80
    },
    {
      "epoch": 10.99,
      "learning_rate": 0.0001,
      "loss": 0.1047,
      "step": 88
    },
    {
      "epoch": 11.99,
      "learning_rate": 0.0001,
      "loss": 0.0988,
      "step": 96
    },
    {
      "epoch": 12.99,
      "learning_rate": 0.0001,
      "loss": 0.0823,
      "step": 104
    },
    {
      "epoch": 13.99,
      "learning_rate": 0.0001,
      "loss": 0.0755,
      "step": 112
    },
    {
      "epoch": 14.99,
      "learning_rate": 0.0001,
      "loss": 0.0697,
      "step": 120
    },
    {
      "epoch": 15.99,
      "learning_rate": 0.0001,
      "loss": 0.061,
      "step": 128
    },
    {
      "epoch": 15.99,
      "eval_exact_match": 0.5114200595829196,
      "eval_exec": 0.6007944389275075,
      "eval_loss": 0.1975870430469513,
      "eval_runtime": 814.4903,
      "eval_samples_per_second": 1.596,
      "step": 128
    },
    {
      "epoch": 16.99,
      "learning_rate": 0.0001,
      "loss": 0.0582,
      "step": 136
    },
    {
      "epoch": 17.99,
      "learning_rate": 0.0001,
      "loss": 0.0551,
      "step": 144
    },
    {
      "epoch": 18.99,
      "learning_rate": 0.0001,
      "loss": 0.053,
      "step": 152
    },
    {
      "epoch": 19.99,
      "learning_rate": 0.0001,
      "loss": 0.0464,
      "step": 160
    },
    {
      "epoch": 20.99,
      "learning_rate": 0.0001,
      "loss": 0.0447,
      "step": 168
    },
    {
      "epoch": 21.99,
      "learning_rate": 0.0001,
      "loss": 0.0476,
      "step": 176
    },
    {
      "epoch": 22.99,
      "learning_rate": 0.0001,
      "loss": 0.037,
      "step": 184
    },
    {
      "epoch": 23.99,
      "learning_rate": 0.0001,
      "loss": 0.0333,
      "step": 192
    },
    {
      "epoch": 23.99,
      "eval_exact_match": 0.506454816285998,
      "eval_exec": 0.6077457795431976,
      "eval_loss": 0.22392722964286804,
      "eval_runtime": 930.363,
      "eval_samples_per_second": 1.397,
      "step": 192
    },
    {
      "epoch": 24.99,
      "learning_rate": 0.0001,
      "loss": 0.0302,
      "step": 200
    },
    {
      "epoch": 25.99,
      "learning_rate": 0.0001,
      "loss": 0.0293,
      "step": 208
    },
    {
      "epoch": 26.99,
      "learning_rate": 0.0001,
      "loss": 0.03,
      "step": 216
    },
    {
      "epoch": 27.99,
      "learning_rate": 0.0001,
      "loss": 0.026,
      "step": 224
    },
    {
      "epoch": 28.99,
      "learning_rate": 0.0001,
      "loss": 0.0228,
      "step": 232
    },
    {
      "epoch": 29.99,
      "learning_rate": 0.0001,
      "loss": 0.0237,
      "step": 240
    },
    {
      "epoch": 30.99,
      "learning_rate": 0.0001,
      "loss": 0.0208,
      "step": 248
    },
    {
      "epoch": 31.99,
      "learning_rate": 0.0001,
      "loss": 0.0191,
      "step": 256
    },
    {
      "epoch": 31.99,
      "eval_exact_match": 0.5362462760675273,
      "eval_exec": 0.6285998013902682,
      "eval_loss": 0.2673013508319855,
      "eval_runtime": 685.3833,
      "eval_samples_per_second": 1.897,
      "step": 256
    },
    {
      "epoch": 32.99,
      "learning_rate": 0.0001,
      "loss": 0.0179,
      "step": 264
    },
    {
      "epoch": 33.99,
      "learning_rate": 0.0001,
      "loss": 0.0172,
      "step": 272
    },
    {
      "epoch": 34.99,
      "learning_rate": 0.0001,
      "loss": 0.0164,
      "step": 280
    },
    {
      "epoch": 35.99,
      "learning_rate": 0.0001,
      "loss": 0.0157,
      "step": 288
    },
    {
      "epoch": 36.99,
      "learning_rate": 0.0001,
      "loss": 0.0178,
      "step": 296
    },
    {
      "epoch": 37.99,
      "learning_rate": 0.0001,
      "loss": 0.0228,
      "step": 304
    },
    {
      "epoch": 38.99,
      "learning_rate": 0.0001,
      "loss": 0.0203,
      "step": 312
    },
    {
      "epoch": 39.99,
      "learning_rate": 0.0001,
      "loss": 0.0165,
      "step": 320
    },
    {
      "epoch": 39.99,
      "eval_exact_match": 0.5183714001986097,
      "eval_exec": 0.6256206554121152,
      "eval_loss": 0.2689046263694763,
      "eval_runtime": 854.2968,
      "eval_samples_per_second": 1.522,
      "step": 320
    },
    {
      "epoch": 40.99,
      "learning_rate": 0.0001,
      "loss": 0.0123,
      "step": 328
    },
    {
      "epoch": 41.99,
      "learning_rate": 0.0001,
      "loss": 0.0107,
      "step": 336
    },
    {
      "epoch": 42.99,
      "learning_rate": 0.0001,
      "loss": 0.0108,
      "step": 344
    },
    {
      "epoch": 43.99,
      "learning_rate": 0.0001,
      "loss": 0.0096,
      "step": 352
    },
    {
      "epoch": 44.99,
      "learning_rate": 0.0001,
      "loss": 0.0089,
      "step": 360
    },
    {
      "epoch": 45.99,
      "learning_rate": 0.0001,
      "loss": 0.0089,
      "step": 368
    },
    {
      "epoch": 46.99,
      "learning_rate": 0.0001,
      "loss": 0.0085,
      "step": 376
    },
    {
      "epoch": 47.99,
      "learning_rate": 0.0001,
      "loss": 0.0072,
      "step": 384
    },
    {
      "epoch": 47.99,
      "eval_exact_match": 0.5283018867924528,
      "eval_exec": 0.6206554121151936,
      "eval_loss": 0.32647866010665894,
      "eval_runtime": 712.6762,
      "eval_samples_per_second": 1.824,
      "step": 384
    },
    {
      "epoch": 48.99,
      "learning_rate": 0.0001,
      "loss": 0.009,
      "step": 392
    },
    {
      "epoch": 49.99,
      "learning_rate": 0.0001,
      "loss": 0.0093,
      "step": 400
    },
    {
      "epoch": 50.99,
      "learning_rate": 0.0001,
      "loss": 0.0072,
      "step": 408
    },
    {
      "epoch": 51.99,
      "learning_rate": 0.0001,
      "loss": 0.0069,
      "step": 416
    },
    {
      "epoch": 52.99,
      "learning_rate": 0.0001,
      "loss": 0.0058,
      "step": 424
    },
    {
      "epoch": 53.99,
      "learning_rate": 0.0001,
      "loss": 0.0062,
      "step": 432
    },
    {
      "epoch": 54.99,
      "learning_rate": 0.0001,
      "loss": 0.0056,
      "step": 440
    },
    {
      "epoch": 55.99,
      "learning_rate": 0.0001,
      "loss": 0.0066,
      "step": 448
    },
    {
      "epoch": 55.99,
      "eval_exact_match": 0.5372393247269116,
      "eval_exec": 0.6206554121151936,
      "eval_loss": 0.3149637281894684,
      "eval_runtime": 877.8005,
      "eval_samples_per_second": 1.481,
      "step": 448
    },
    {
      "epoch": 56.99,
      "learning_rate": 0.0001,
      "loss": 0.0072,
      "step": 456
    },
    {
      "epoch": 57.99,
      "learning_rate": 0.0001,
      "loss": 0.0086,
      "step": 464
    },
    {
      "epoch": 58.99,
      "learning_rate": 0.0001,
      "loss": 0.005,
      "step": 472
    },
    {
      "epoch": 59.99,
      "learning_rate": 0.0001,
      "loss": 0.0047,
      "step": 480
    },
    {
      "epoch": 60.99,
      "learning_rate": 0.0001,
      "loss": 0.0044,
      "step": 488
    },
    {
      "epoch": 61.99,
      "learning_rate": 0.0001,
      "loss": 0.0045,
      "step": 496
    },
    {
      "epoch": 62.99,
      "learning_rate": 0.0001,
      "loss": 0.0068,
      "step": 504
    },
    {
      "epoch": 63.99,
      "learning_rate": 0.0001,
      "loss": 0.0044,
      "step": 512
    },
    {
      "epoch": 63.99,
      "eval_exact_match": 0.5590863952333665,
      "eval_exec": 0.637537239324727,
      "eval_loss": 0.38160499930381775,
      "eval_runtime": 805.0905,
      "eval_samples_per_second": 1.615,
      "step": 512
    },
    {
      "epoch": 64.99,
      "learning_rate": 0.0001,
      "loss": 0.004,
      "step": 520
    },
    {
      "epoch": 65.99,
      "learning_rate": 0.0001,
      "loss": 0.0044,
      "step": 528
    },
    {
      "epoch": 66.99,
      "learning_rate": 0.0001,
      "loss": 0.0038,
      "step": 536
    },
    {
      "epoch": 67.99,
      "learning_rate": 0.0001,
      "loss": 0.0048,
      "step": 544
    },
    {
      "epoch": 68.99,
      "learning_rate": 0.0001,
      "loss": 0.0036,
      "step": 552
    },
    {
      "epoch": 69.99,
      "learning_rate": 0.0001,
      "loss": 0.0033,
      "step": 560
    },
    {
      "epoch": 70.99,
      "learning_rate": 0.0001,
      "loss": 0.0035,
      "step": 568
    },
    {
      "epoch": 71.99,
      "learning_rate": 0.0001,
      "loss": 0.0035,
      "step": 576
    },
    {
      "epoch": 71.99,
      "eval_exact_match": 0.5610724925521351,
      "eval_exec": 0.631578947368421,
      "eval_loss": 0.3941967189311981,
      "eval_runtime": 731.2157,
      "eval_samples_per_second": 1.778,
      "step": 576
    }
  ],
  "max_steps": 24576,
  "num_train_epochs": 3072,
  "total_flos": 4.452475717891916e+18,
  "trial_name": null,
  "trial_params": null
}