{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.95603827256271,
  "global_step": 38500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.13,
      "learning_rate": 1.9741401603310062e-05,
      "loss": 1.9914,
      "step": 500
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.948280320662012e-05,
      "loss": 1.606,
      "step": 1000
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.922420480993018e-05,
      "loss": 1.4903,
      "step": 1500
    },
    {
      "epoch": 0.52,
      "learning_rate": 1.896560641324024e-05,
      "loss": 1.4065,
      "step": 2000
    },
    {
      "epoch": 0.65,
      "learning_rate": 1.87070080165503e-05,
      "loss": 1.3202,
      "step": 2500
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.844840961986036e-05,
      "loss": 1.288,
      "step": 3000
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.8189811223170417e-05,
      "loss": 1.2592,
      "step": 3500
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.1307439804077148,
      "eval_runtime": 12.893,
      "eval_samples_per_second": 1193.128,
      "eval_steps_per_second": 149.151,
      "step": 3867
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.7931212826480477e-05,
      "loss": 1.2107,
      "step": 4000
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.7672614429790537e-05,
      "loss": 1.1851,
      "step": 4500
    },
    {
      "epoch": 1.29,
      "learning_rate": 1.7414016033100598e-05,
      "loss": 1.1513,
      "step": 5000
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.7155417636410658e-05,
      "loss": 1.1181,
      "step": 5500
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.6896819239720715e-05,
      "loss": 1.0948,
      "step": 6000
    },
    {
      "epoch": 1.68,
      "learning_rate": 1.6638220843030775e-05,
      "loss": 1.0955,
      "step": 6500
    },
    {
      "epoch": 1.81,
      "learning_rate": 1.6379622446340835e-05,
      "loss": 1.0562,
      "step": 7000
    },
    {
      "epoch": 1.94,
      "learning_rate": 1.6121024049650892e-05,
      "loss": 1.0578,
      "step": 7500
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.9538503885269165,
      "eval_runtime": 12.8064,
      "eval_samples_per_second": 1201.199,
      "eval_steps_per_second": 150.16,
      "step": 7734
    },
    {
      "epoch": 2.07,
      "learning_rate": 1.5862425652960952e-05,
      "loss": 1.0484,
      "step": 8000
    },
    {
      "epoch": 2.2,
      "learning_rate": 1.5603827256271013e-05,
      "loss": 1.0226,
      "step": 8500
    },
    {
      "epoch": 2.33,
      "learning_rate": 1.5345228859581073e-05,
      "loss": 1.0189,
      "step": 9000
    },
    {
      "epoch": 2.46,
      "learning_rate": 1.5086630462891131e-05,
      "loss": 0.9812,
      "step": 9500
    },
    {
      "epoch": 2.59,
      "learning_rate": 1.482803206620119e-05,
      "loss": 0.9543,
      "step": 10000
    },
    {
      "epoch": 2.72,
      "learning_rate": 1.456943366951125e-05,
      "loss": 0.9917,
      "step": 10500
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.431083527282131e-05,
      "loss": 0.9423,
      "step": 11000
    },
    {
      "epoch": 2.97,
      "learning_rate": 1.4052236876131369e-05,
      "loss": 0.9628,
      "step": 11500
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.8621994256973267,
      "eval_runtime": 12.1534,
      "eval_samples_per_second": 1265.738,
      "eval_steps_per_second": 158.228,
      "step": 11601
    },
    {
      "epoch": 3.1,
      "learning_rate": 1.379363847944143e-05,
      "loss": 0.9401,
      "step": 12000
    },
    {
      "epoch": 3.23,
      "learning_rate": 1.3535040082751488e-05,
      "loss": 0.9208,
      "step": 12500
    },
    {
      "epoch": 3.36,
      "learning_rate": 1.3276441686061548e-05,
      "loss": 0.9068,
      "step": 13000
    },
    {
      "epoch": 3.49,
      "learning_rate": 1.3017843289371609e-05,
      "loss": 0.9114,
      "step": 13500
    },
    {
      "epoch": 3.62,
      "learning_rate": 1.2759244892681665e-05,
      "loss": 0.9146,
      "step": 14000
    },
    {
      "epoch": 3.75,
      "learning_rate": 1.2500646495991726e-05,
      "loss": 0.8964,
      "step": 14500
    },
    {
      "epoch": 3.88,
      "learning_rate": 1.2242048099301784e-05,
      "loss": 0.9084,
      "step": 15000
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.8124715685844421,
      "eval_runtime": 12.1415,
      "eval_samples_per_second": 1266.974,
      "eval_steps_per_second": 158.382,
      "step": 15468
    },
    {
      "epoch": 4.01,
      "learning_rate": 1.1983449702611844e-05,
      "loss": 0.8739,
      "step": 15500
    },
    {
      "epoch": 4.14,
      "learning_rate": 1.1724851305921905e-05,
      "loss": 0.8625,
      "step": 16000
    },
    {
      "epoch": 4.27,
      "learning_rate": 1.1466252909231963e-05,
      "loss": 0.8507,
      "step": 16500
    },
    {
      "epoch": 4.4,
      "learning_rate": 1.1207654512542024e-05,
      "loss": 0.8759,
      "step": 17000
    },
    {
      "epoch": 4.53,
      "learning_rate": 1.0949056115852084e-05,
      "loss": 0.8383,
      "step": 17500
    },
    {
      "epoch": 4.65,
      "learning_rate": 1.0690457719162142e-05,
      "loss": 0.8487,
      "step": 18000
    },
    {
      "epoch": 4.78,
      "learning_rate": 1.0431859322472203e-05,
      "loss": 0.8595,
      "step": 18500
    },
    {
      "epoch": 4.91,
      "learning_rate": 1.017326092578226e-05,
      "loss": 0.8374,
      "step": 19000
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.7769160270690918,
      "eval_runtime": 12.8439,
      "eval_samples_per_second": 1197.691,
      "eval_steps_per_second": 149.721,
      "step": 19335
    },
    {
      "epoch": 5.04,
      "learning_rate": 9.91466252909232e-06,
      "loss": 0.8352,
      "step": 19500
    },
    {
      "epoch": 5.17,
      "learning_rate": 9.65606413240238e-06,
      "loss": 0.8663,
      "step": 20000
    },
    {
      "epoch": 5.3,
      "learning_rate": 9.397465735712439e-06,
      "loss": 0.8169,
      "step": 20500
    },
    {
      "epoch": 5.43,
      "learning_rate": 9.138867339022499e-06,
      "loss": 0.8187,
      "step": 21000
    },
    {
      "epoch": 5.56,
      "learning_rate": 8.880268942332558e-06,
      "loss": 0.8153,
      "step": 21500
    },
    {
      "epoch": 5.69,
      "learning_rate": 8.621670545642618e-06,
      "loss": 0.8133,
      "step": 22000
    },
    {
      "epoch": 5.82,
      "learning_rate": 8.363072148952676e-06,
      "loss": 0.7901,
      "step": 22500
    },
    {
      "epoch": 5.95,
      "learning_rate": 8.104473752262737e-06,
      "loss": 0.8139,
      "step": 23000
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.7311471104621887,
      "eval_runtime": 12.8348,
      "eval_samples_per_second": 1198.535,
      "eval_steps_per_second": 149.827,
      "step": 23202
    },
    {
      "epoch": 6.08,
      "learning_rate": 7.845875355572797e-06,
      "loss": 0.7811,
      "step": 23500
    },
    {
      "epoch": 6.21,
      "learning_rate": 7.5872769588828555e-06,
      "loss": 0.8063,
      "step": 24000
    },
    {
      "epoch": 6.34,
      "learning_rate": 7.328678562192915e-06,
      "loss": 0.785,
      "step": 24500
    },
    {
      "epoch": 6.46,
      "learning_rate": 7.070080165502975e-06,
      "loss": 0.7909,
      "step": 25000
    },
    {
      "epoch": 6.59,
      "learning_rate": 6.811481768813034e-06,
      "loss": 0.7727,
      "step": 25500
    },
    {
      "epoch": 6.72,
      "learning_rate": 6.552883372123093e-06,
      "loss": 0.7779,
      "step": 26000
    },
    {
      "epoch": 6.85,
      "learning_rate": 6.294284975433153e-06,
      "loss": 0.7627,
      "step": 26500
    },
    {
      "epoch": 6.98,
      "learning_rate": 6.035686578743212e-06,
      "loss": 0.7577,
      "step": 27000
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.6985681653022766,
      "eval_runtime": 12.1148,
      "eval_samples_per_second": 1269.772,
      "eval_steps_per_second": 158.732,
      "step": 27069
    },
    {
      "epoch": 7.11,
      "learning_rate": 5.777088182053272e-06,
      "loss": 0.7761,
      "step": 27500
    },
    {
      "epoch": 7.24,
      "learning_rate": 5.518489785363331e-06,
      "loss": 0.7782,
      "step": 28000
    },
    {
      "epoch": 7.37,
      "learning_rate": 5.25989138867339e-06,
      "loss": 0.7335,
      "step": 28500
    },
    {
      "epoch": 7.5,
      "learning_rate": 5.00129299198345e-06,
      "loss": 0.7656,
      "step": 29000
    },
    {
      "epoch": 7.63,
      "learning_rate": 4.742694595293509e-06,
      "loss": 0.761,
      "step": 29500
    },
    {
      "epoch": 7.76,
      "learning_rate": 4.484096198603569e-06,
      "loss": 0.7517,
      "step": 30000
    },
    {
      "epoch": 7.89,
      "learning_rate": 4.225497801913628e-06,
      "loss": 0.75,
      "step": 30500
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.6888388991355896,
      "eval_runtime": 12.1237,
      "eval_samples_per_second": 1268.839,
      "eval_steps_per_second": 158.615,
      "step": 30936
    },
    {
      "epoch": 8.02,
      "learning_rate": 3.966899405223688e-06,
      "loss": 0.7545,
      "step": 31000
    },
    {
      "epoch": 8.15,
      "learning_rate": 3.708301008533747e-06,
      "loss": 0.723,
      "step": 31500
    },
    {
      "epoch": 8.28,
      "learning_rate": 3.449702611843807e-06,
      "loss": 0.7352,
      "step": 32000
    },
    {
      "epoch": 8.4,
      "learning_rate": 3.191104215153866e-06,
      "loss": 0.7253,
      "step": 32500
    },
    {
      "epoch": 8.53,
      "learning_rate": 2.932505818463926e-06,
      "loss": 0.7248,
      "step": 33000
    },
    {
      "epoch": 8.66,
      "learning_rate": 2.6739074217739853e-06,
      "loss": 0.7371,
      "step": 33500
    },
    {
      "epoch": 8.79,
      "learning_rate": 2.4153090250840447e-06,
      "loss": 0.7284,
      "step": 34000
    },
    {
      "epoch": 8.92,
      "learning_rate": 2.156710628394104e-06,
      "loss": 0.7205,
      "step": 34500
    },
    {
      "epoch": 9.0,
      "eval_loss": 0.6802728772163391,
      "eval_runtime": 12.1115,
      "eval_samples_per_second": 1270.111,
      "eval_steps_per_second": 158.774,
      "step": 34803
    },
    {
      "epoch": 9.05,
      "learning_rate": 1.8981122317041636e-06,
      "loss": 0.7059,
      "step": 35000
    },
    {
      "epoch": 9.18,
      "learning_rate": 1.6395138350142232e-06,
      "loss": 0.7321,
      "step": 35500
    },
    {
      "epoch": 9.31,
      "learning_rate": 1.3809154383242826e-06,
      "loss": 0.7392,
      "step": 36000
    },
    {
      "epoch": 9.44,
      "learning_rate": 1.122317041634342e-06,
      "loss": 0.7314,
      "step": 36500
    },
    {
      "epoch": 9.57,
      "learning_rate": 8.637186449444015e-07,
      "loss": 0.73,
      "step": 37000
    },
    {
      "epoch": 9.7,
      "learning_rate": 6.051202482544609e-07,
      "loss": 0.7199,
      "step": 37500
    },
    {
      "epoch": 9.83,
      "learning_rate": 3.465218515645203e-07,
      "loss": 0.7258,
      "step": 38000
    },
    {
      "epoch": 9.96,
      "learning_rate": 8.792345487457979e-08,
      "loss": 0.7562,
      "step": 38500
    }
  ],
  "max_steps": 38670,
  "num_train_epochs": 10,
  "total_flos": 1994099670728550.0,
  "trial_name": null,
  "trial_params": null
}