{
  "best_metric": 0.4246249496936798,
  "best_model_checkpoint": "OutModelPolicy/checkpoint-50000",
  "epoch": 0.699623602501854,
  "global_step": 50000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 1.986035512894063e-05,
      "loss": 0.5152,
      "step": 500
    },
    {
      "epoch": 0.01,
      "learning_rate": 1.972043040844026e-05,
      "loss": 0.4835,
      "step": 1000
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.958050568793989e-05,
      "loss": 0.4742,
      "step": 1500
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.944058096743952e-05,
      "loss": 0.4763,
      "step": 2000
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.930065624693915e-05,
      "loss": 0.464,
      "step": 2500
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.916101137587978e-05,
      "loss": 0.4635,
      "step": 3000
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.902108665537941e-05,
      "loss": 0.4621,
      "step": 3500
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.8881161934879037e-05,
      "loss": 0.4525,
      "step": 4000
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.8741237214378666e-05,
      "loss": 0.4636,
      "step": 4500
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.8601872192760296e-05,
      "loss": 0.4491,
      "step": 5000
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.8461947472259925e-05,
      "loss": 0.4502,
      "step": 5500
    },
    {
      "epoch": 0.08,
      "learning_rate": 1.8322022751759554e-05,
      "loss": 0.4512,
      "step": 6000
    },
    {
      "epoch": 0.09,
      "learning_rate": 1.8182377880700184e-05,
      "loss": 0.4546,
      "step": 6500
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.8042453160199812e-05,
      "loss": 0.4477,
      "step": 7000
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.790252843969944e-05,
      "loss": 0.4582,
      "step": 7500
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.7762603719199074e-05,
      "loss": 0.4452,
      "step": 8000
    },
    {
      "epoch": 0.12,
      "learning_rate": 1.7622678998698702e-05,
      "loss": 0.4491,
      "step": 8500
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.748275427819833e-05,
      "loss": 0.4511,
      "step": 9000
    },
    {
      "epoch": 0.13,
      "learning_rate": 1.734282955769796e-05,
      "loss": 0.442,
      "step": 9500
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.720290483719759e-05,
      "loss": 0.4403,
      "step": 10000
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.7062980116697218e-05,
      "loss": 0.4475,
      "step": 10500
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.6923335245637847e-05,
      "loss": 0.4409,
      "step": 11000
    },
    {
      "epoch": 0.16,
      "learning_rate": 1.678341052513748e-05,
      "loss": 0.4427,
      "step": 11500
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.6643485804637108e-05,
      "loss": 0.4335,
      "step": 12000
    },
    {
      "epoch": 0.17,
      "learning_rate": 1.6503561084136737e-05,
      "loss": 0.4474,
      "step": 12500
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.6363636363636366e-05,
      "loss": 0.4552,
      "step": 13000
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.6223711643135995e-05,
      "loss": 0.4405,
      "step": 13500
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.6083786922635624e-05,
      "loss": 0.4331,
      "step": 14000
    },
    {
      "epoch": 0.2,
      "learning_rate": 1.5943862202135252e-05,
      "loss": 0.4338,
      "step": 14500
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.5804217331075882e-05,
      "loss": 0.4339,
      "step": 15000
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.566429261057551e-05,
      "loss": 0.4355,
      "step": 15500
    },
    {
      "epoch": 0.22,
      "learning_rate": 1.552436789007514e-05,
      "loss": 0.4466,
      "step": 16000
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.538444316957477e-05,
      "loss": 0.4368,
      "step": 16500
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.5244518449074399e-05,
      "loss": 0.4343,
      "step": 17000
    },
    {
      "epoch": 0.24,
      "learning_rate": 1.5104873578015029e-05,
      "loss": 0.4404,
      "step": 17500
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.4964948857514659e-05,
      "loss": 0.4367,
      "step": 18000
    },
    {
      "epoch": 0.26,
      "learning_rate": 1.4825024137014288e-05,
      "loss": 0.4389,
      "step": 18500
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.4685099416513917e-05,
      "loss": 0.4396,
      "step": 19000
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.4545454545454546e-05,
      "loss": 0.4364,
      "step": 19500
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.4405529824954177e-05,
      "loss": 0.4434,
      "step": 20000
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.4265605104453806e-05,
      "loss": 0.4432,
      "step": 20500
    },
    {
      "epoch": 0.29,
      "learning_rate": 1.4125680383953434e-05,
      "loss": 0.4459,
      "step": 21000
    },
    {
      "epoch": 0.3,
      "learning_rate": 1.3986035512894066e-05,
      "loss": 0.4415,
      "step": 21500
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.3846110792393695e-05,
      "loss": 0.4373,
      "step": 22000
    },
    {
      "epoch": 0.31,
      "learning_rate": 1.3706465921334324e-05,
      "loss": 0.439,
      "step": 22500
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.3566541200833951e-05,
      "loss": 0.4331,
      "step": 23000
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.342661648033358e-05,
      "loss": 0.4361,
      "step": 23500
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.328669175983321e-05,
      "loss": 0.4347,
      "step": 24000
    },
    {
      "epoch": 0.34,
      "learning_rate": 1.314676703933284e-05,
      "loss": 0.4319,
      "step": 24500
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.3006842318832468e-05,
      "loss": 0.4321,
      "step": 25000
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.2866917598332099e-05,
      "loss": 0.4208,
      "step": 25500
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.2727272727272728e-05,
      "loss": 0.4404,
      "step": 26000
    },
    {
      "epoch": 0.37,
      "learning_rate": 1.2587348006772357e-05,
      "loss": 0.4248,
      "step": 26500
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.2447423286271986e-05,
      "loss": 0.4351,
      "step": 27000
    },
    {
      "epoch": 0.38,
      "learning_rate": 1.2307498565771617e-05,
      "loss": 0.4302,
      "step": 27500
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.2167573845271245e-05,
      "loss": 0.4335,
      "step": 28000
    },
    {
      "epoch": 0.4,
      "learning_rate": 1.2027649124770874e-05,
      "loss": 0.4286,
      "step": 28500
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.1887724404270505e-05,
      "loss": 0.4228,
      "step": 29000
    },
    {
      "epoch": 0.41,
      "learning_rate": 1.1748079533211134e-05,
      "loss": 0.4336,
      "step": 29500
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.1608154812710763e-05,
      "loss": 0.4313,
      "step": 30000
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.1468230092210392e-05,
      "loss": 0.4322,
      "step": 30500
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.1328305371710019e-05,
      "loss": 0.4439,
      "step": 31000
    },
    {
      "epoch": 0.44,
      "learning_rate": 1.118866050065065e-05,
      "loss": 0.4354,
      "step": 31500
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.104873578015028e-05,
      "loss": 0.4268,
      "step": 32000
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.0908811059649908e-05,
      "loss": 0.4281,
      "step": 32500
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.0768886339149539e-05,
      "loss": 0.4261,
      "step": 33000
    },
    {
      "epoch": 0.47,
      "learning_rate": 1.0628961618649167e-05,
      "loss": 0.4272,
      "step": 33500
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.0489316747589797e-05,
      "loss": 0.4261,
      "step": 34000
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.0349392027089426e-05,
      "loss": 0.4345,
      "step": 34500
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.0209467306589056e-05,
      "loss": 0.4292,
      "step": 35000
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.0069542586088685e-05,
      "loss": 0.4323,
      "step": 35500
    },
    {
      "epoch": 0.5,
      "learning_rate": 9.929617865588314e-06,
      "loss": 0.4303,
      "step": 36000
    },
    {
      "epoch": 0.51,
      "learning_rate": 9.789693145087945e-06,
      "loss": 0.425,
      "step": 36500
    },
    {
      "epoch": 0.52,
      "learning_rate": 9.649768424587572e-06,
      "loss": 0.4297,
      "step": 37000
    },
    {
      "epoch": 0.52,
      "learning_rate": 9.509843704087202e-06,
      "loss": 0.4262,
      "step": 37500
    },
    {
      "epoch": 0.53,
      "learning_rate": 9.370198833027832e-06,
      "loss": 0.431,
      "step": 38000
    },
    {
      "epoch": 0.54,
      "learning_rate": 9.230553961968461e-06,
      "loss": 0.4225,
      "step": 38500
    },
    {
      "epoch": 0.55,
      "learning_rate": 9.09062924146809e-06,
      "loss": 0.44,
      "step": 39000
    },
    {
      "epoch": 0.55,
      "learning_rate": 8.95070452096772e-06,
      "loss": 0.4229,
      "step": 39500
    },
    {
      "epoch": 0.56,
      "learning_rate": 8.81077980046735e-06,
      "loss": 0.421,
      "step": 40000
    },
    {
      "epoch": 0.57,
      "learning_rate": 8.670855079966978e-06,
      "loss": 0.4366,
      "step": 40500
    },
    {
      "epoch": 0.57,
      "learning_rate": 8.530930359466609e-06,
      "loss": 0.424,
      "step": 41000
    },
    {
      "epoch": 0.58,
      "learning_rate": 8.391005638966236e-06,
      "loss": 0.4236,
      "step": 41500
    },
    {
      "epoch": 0.59,
      "learning_rate": 8.251080918465867e-06,
      "loss": 0.4222,
      "step": 42000
    },
    {
      "epoch": 0.59,
      "learning_rate": 8.111436047406496e-06,
      "loss": 0.4187,
      "step": 42500
    },
    {
      "epoch": 0.6,
      "learning_rate": 7.971511326906125e-06,
      "loss": 0.4444,
      "step": 43000
    },
    {
      "epoch": 0.61,
      "learning_rate": 7.831586606405754e-06,
      "loss": 0.4278,
      "step": 43500
    },
    {
      "epoch": 0.62,
      "learning_rate": 7.691661885905384e-06,
      "loss": 0.431,
      "step": 44000
    },
    {
      "epoch": 0.62,
      "learning_rate": 7.552017014846014e-06,
      "loss": 0.4258,
      "step": 44500
    },
    {
      "epoch": 0.63,
      "learning_rate": 7.4123721437866435e-06,
      "loss": 0.4252,
      "step": 45000
    },
    {
      "epoch": 0.64,
      "learning_rate": 7.272447423286273e-06,
      "loss": 0.4279,
      "step": 45500
    },
    {
      "epoch": 0.64,
      "learning_rate": 7.132522702785902e-06,
      "loss": 0.4279,
      "step": 46000
    },
    {
      "epoch": 0.65,
      "learning_rate": 6.992597982285531e-06,
      "loss": 0.4178,
      "step": 46500
    },
    {
      "epoch": 0.66,
      "learning_rate": 6.85267326178516e-06,
      "loss": 0.4201,
      "step": 47000
    },
    {
      "epoch": 0.66,
      "learning_rate": 6.71302839072579e-06,
      "loss": 0.4244,
      "step": 47500
    },
    {
      "epoch": 0.67,
      "learning_rate": 6.573103670225419e-06,
      "loss": 0.4169,
      "step": 48000
    },
    {
      "epoch": 0.68,
      "learning_rate": 6.433178949725049e-06,
      "loss": 0.427,
      "step": 48500
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.293254229224678e-06,
      "loss": 0.4325,
      "step": 49000
    },
    {
      "epoch": 0.69,
      "learning_rate": 6.153329508724307e-06,
      "loss": 0.4206,
      "step": 49500
    },
    {
      "epoch": 0.7,
      "learning_rate": 6.013404788223935e-06,
      "loss": 0.4342,
      "step": 50000
    },
    {
      "epoch": 0.7,
      "eval_accuracy": 0.807313903842584,
      "eval_loss": 0.4246249496936798,
      "eval_runtime": 59.7232,
      "eval_samples_per_second": 90.199,
      "eval_steps_per_second": 6.028,
      "step": 50000
    }
  ],
  "max_steps": 71467,
  "num_train_epochs": 1,
  "total_flos": 9.090578558220155e+16,
  "trial_name": null,
  "trial_params": null
}