{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 50.0,
  "eval_steps": 500,
  "global_step": 7400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.22874597126094132,
      "eval_loss": 5.837828636169434,
      "eval_runtime": 19.7897,
      "eval_samples_per_second": 59.071,
      "eval_steps_per_second": 1.87,
      "step": 148
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.24914672818864436,
      "eval_loss": 5.573156356811523,
      "eval_runtime": 19.7414,
      "eval_samples_per_second": 59.216,
      "eval_steps_per_second": 1.874,
      "step": 296
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.26800943567410634,
      "eval_loss": 5.379988670349121,
      "eval_runtime": 19.7422,
      "eval_samples_per_second": 59.213,
      "eval_steps_per_second": 1.874,
      "step": 444
    },
    {
      "epoch": 3.38,
      "learning_rate": 9.324324324324325e-06,
      "loss": 5.761,
      "step": 500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.28418487699924827,
      "eval_loss": 5.225850582122803,
      "eval_runtime": 19.7377,
      "eval_samples_per_second": 59.227,
      "eval_steps_per_second": 1.875,
      "step": 592
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.2963078172659011,
      "eval_loss": 5.1006011962890625,
      "eval_runtime": 19.742,
      "eval_samples_per_second": 59.214,
      "eval_steps_per_second": 1.874,
      "step": 740
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.30673717500064807,
      "eval_loss": 4.999685764312744,
      "eval_runtime": 19.749,
      "eval_samples_per_second": 59.193,
      "eval_steps_per_second": 1.874,
      "step": 888
    },
    {
      "epoch": 6.76,
      "learning_rate": 8.64864864864865e-06,
      "loss": 5.1725,
      "step": 1000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.3152828542050099,
      "eval_loss": 4.910343170166016,
      "eval_runtime": 19.7422,
      "eval_samples_per_second": 59.213,
      "eval_steps_per_second": 1.874,
      "step": 1036
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.322955819961808,
      "eval_loss": 4.828683853149414,
      "eval_runtime": 19.7403,
      "eval_samples_per_second": 59.219,
      "eval_steps_per_second": 1.874,
      "step": 1184
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.3300757791775756,
      "eval_loss": 4.757791996002197,
      "eval_runtime": 19.7469,
      "eval_samples_per_second": 59.199,
      "eval_steps_per_second": 1.874,
      "step": 1332
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.3375845711175052,
      "eval_loss": 4.694174766540527,
      "eval_runtime": 19.7442,
      "eval_samples_per_second": 59.207,
      "eval_steps_per_second": 1.874,
      "step": 1480
    },
    {
      "epoch": 10.14,
      "learning_rate": 7.972972972972974e-06,
      "loss": 4.8482,
      "step": 1500
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.3439355056121523,
      "eval_loss": 4.63644552230835,
      "eval_runtime": 19.7467,
      "eval_samples_per_second": 59.2,
      "eval_steps_per_second": 1.874,
      "step": 1628
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.34968158920254727,
      "eval_loss": 4.5813164710998535,
      "eval_runtime": 19.7421,
      "eval_samples_per_second": 59.214,
      "eval_steps_per_second": 1.874,
      "step": 1776
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.3554363135201459,
      "eval_loss": 4.532833099365234,
      "eval_runtime": 19.7462,
      "eval_samples_per_second": 59.201,
      "eval_steps_per_second": 1.874,
      "step": 1924
    },
    {
      "epoch": 13.51,
      "learning_rate": 7.297297297297298e-06,
      "loss": 4.609,
      "step": 2000
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.3610182232936724,
      "eval_loss": 4.4896931648254395,
      "eval_runtime": 19.7471,
      "eval_samples_per_second": 59.199,
      "eval_steps_per_second": 1.874,
      "step": 2072
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.3656582938020064,
      "eval_loss": 4.4454345703125,
      "eval_runtime": 19.7467,
      "eval_samples_per_second": 59.2,
      "eval_steps_per_second": 1.874,
      "step": 2220
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.37016011267508275,
      "eval_loss": 4.413242340087891,
      "eval_runtime": 19.7438,
      "eval_samples_per_second": 59.208,
      "eval_steps_per_second": 1.874,
      "step": 2368
    },
    {
      "epoch": 16.89,
      "learning_rate": 6.621621621621622e-06,
      "loss": 4.4241,
      "step": 2500
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.3737978588277989,
      "eval_loss": 4.374161243438721,
      "eval_runtime": 19.7484,
      "eval_samples_per_second": 59.195,
      "eval_steps_per_second": 1.874,
      "step": 2516
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.3782132704288393,
      "eval_loss": 4.343791961669922,
      "eval_runtime": 19.7509,
      "eval_samples_per_second": 59.187,
      "eval_steps_per_second": 1.873,
      "step": 2664
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.38171276494629786,
      "eval_loss": 4.316366672515869,
      "eval_runtime": 19.7486,
      "eval_samples_per_second": 59.194,
      "eval_steps_per_second": 1.874,
      "step": 2812
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.3847715823763728,
      "eval_loss": 4.2879157066345215,
      "eval_runtime": 19.7502,
      "eval_samples_per_second": 59.189,
      "eval_steps_per_second": 1.873,
      "step": 2960
    },
    {
      "epoch": 20.27,
      "learning_rate": 5.945945945945947e-06,
      "loss": 4.283,
      "step": 3000
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.3878131183520405,
      "eval_loss": 4.260156631469727,
      "eval_runtime": 19.7607,
      "eval_samples_per_second": 59.158,
      "eval_steps_per_second": 1.872,
      "step": 3108
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.39024116269625253,
      "eval_loss": 4.23726224899292,
      "eval_runtime": 19.749,
      "eval_samples_per_second": 59.193,
      "eval_steps_per_second": 1.874,
      "step": 3256
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.3931530877638662,
      "eval_loss": 4.216041088104248,
      "eval_runtime": 19.7399,
      "eval_samples_per_second": 59.22,
      "eval_steps_per_second": 1.874,
      "step": 3404
    },
    {
      "epoch": 23.65,
      "learning_rate": 5.2702702702702705e-06,
      "loss": 4.1606,
      "step": 3500
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.39539103610959897,
      "eval_loss": 4.196435928344727,
      "eval_runtime": 19.7434,
      "eval_samples_per_second": 59.21,
      "eval_steps_per_second": 1.874,
      "step": 3552
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.39759442154651736,
      "eval_loss": 4.181632041931152,
      "eval_runtime": 19.7448,
      "eval_samples_per_second": 59.205,
      "eval_steps_per_second": 1.874,
      "step": 3700
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.39951266298571686,
      "eval_loss": 4.160492897033691,
      "eval_runtime": 19.7431,
      "eval_samples_per_second": 59.21,
      "eval_steps_per_second": 1.874,
      "step": 3848
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.4016382818778028,
      "eval_loss": 4.144294738769531,
      "eval_runtime": 19.7428,
      "eval_samples_per_second": 59.212,
      "eval_steps_per_second": 1.874,
      "step": 3996
    },
    {
      "epoch": 27.03,
      "learning_rate": 4.594594594594596e-06,
      "loss": 4.0685,
      "step": 4000
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.4034355531361519,
      "eval_loss": 4.129027366638184,
      "eval_runtime": 19.7497,
      "eval_samples_per_second": 59.191,
      "eval_steps_per_second": 1.873,
      "step": 4144
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.40515505784966865,
      "eval_loss": 4.114578723907471,
      "eval_runtime": 19.7461,
      "eval_samples_per_second": 59.202,
      "eval_steps_per_second": 1.874,
      "step": 4292
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.40673631092792767,
      "eval_loss": 4.103656768798828,
      "eval_runtime": 19.7473,
      "eval_samples_per_second": 59.198,
      "eval_steps_per_second": 1.874,
      "step": 4440
    },
    {
      "epoch": 30.41,
      "learning_rate": 3.918918918918919e-06,
      "loss": 3.9859,
      "step": 4500
    },
    {
      "epoch": 31.0,
      "eval_accuracy": 0.40824843818855794,
      "eval_loss": 4.087373733520508,
      "eval_runtime": 19.744,
      "eval_samples_per_second": 59.208,
      "eval_steps_per_second": 1.874,
      "step": 4588
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.4098124098124098,
      "eval_loss": 4.078439235687256,
      "eval_runtime": 19.7428,
      "eval_samples_per_second": 59.211,
      "eval_steps_per_second": 1.874,
      "step": 4736
    },
    {
      "epoch": 33.0,
      "eval_accuracy": 0.4114714294355013,
      "eval_loss": 4.066919803619385,
      "eval_runtime": 19.7485,
      "eval_samples_per_second": 59.194,
      "eval_steps_per_second": 1.874,
      "step": 4884
    },
    {
      "epoch": 33.78,
      "learning_rate": 3.2432432432432437e-06,
      "loss": 3.9275,
      "step": 5000
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.41251695742713707,
      "eval_loss": 4.058130264282227,
      "eval_runtime": 19.7418,
      "eval_samples_per_second": 59.215,
      "eval_steps_per_second": 1.874,
      "step": 5032
    },
    {
      "epoch": 35.0,
      "eval_accuracy": 0.41355384469156925,
      "eval_loss": 4.047876834869385,
      "eval_runtime": 19.7402,
      "eval_samples_per_second": 59.219,
      "eval_steps_per_second": 1.874,
      "step": 5180
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.41476354650007347,
      "eval_loss": 4.038449287414551,
      "eval_runtime": 19.7398,
      "eval_samples_per_second": 59.22,
      "eval_steps_per_second": 1.874,
      "step": 5328
    },
    {
      "epoch": 37.0,
      "eval_accuracy": 0.4158868410365416,
      "eval_loss": 4.033016204833984,
      "eval_runtime": 19.7451,
      "eval_samples_per_second": 59.204,
      "eval_steps_per_second": 1.874,
      "step": 5476
    },
    {
      "epoch": 37.16,
      "learning_rate": 2.5675675675675675e-06,
      "loss": 3.8799,
      "step": 5500
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 0.41657809921282973,
      "eval_loss": 4.026158332824707,
      "eval_runtime": 19.7412,
      "eval_samples_per_second": 59.216,
      "eval_steps_per_second": 1.874,
      "step": 5624
    },
    {
      "epoch": 39.0,
      "eval_accuracy": 0.4173989682971719,
      "eval_loss": 4.021160125732422,
      "eval_runtime": 19.7395,
      "eval_samples_per_second": 59.221,
      "eval_steps_per_second": 1.874,
      "step": 5772
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.418003819201424,
      "eval_loss": 4.013641834259033,
      "eval_runtime": 19.7429,
      "eval_samples_per_second": 59.211,
      "eval_steps_per_second": 1.874,
      "step": 5920
    },
    {
      "epoch": 40.54,
      "learning_rate": 1.8918918918918922e-06,
      "loss": 3.8348,
      "step": 6000
    },
    {
      "epoch": 41.0,
      "eval_accuracy": 0.4185568257424545,
      "eval_loss": 4.011115550994873,
      "eval_runtime": 19.7447,
      "eval_samples_per_second": 59.206,
      "eval_steps_per_second": 1.874,
      "step": 6068
    },
    {
      "epoch": 42.0,
      "eval_accuracy": 0.4192567246459462,
      "eval_loss": 4.004785060882568,
      "eval_runtime": 19.7414,
      "eval_samples_per_second": 59.216,
      "eval_steps_per_second": 1.874,
      "step": 6216
    },
    {
      "epoch": 43.0,
      "eval_accuracy": 0.41952458718925784,
      "eval_loss": 4.000365257263184,
      "eval_runtime": 19.7443,
      "eval_samples_per_second": 59.207,
      "eval_steps_per_second": 1.874,
      "step": 6364
    },
    {
      "epoch": 43.92,
      "learning_rate": 1.2162162162162164e-06,
      "loss": 3.8151,
      "step": 6500
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.4198615755501983,
      "eval_loss": 3.997826337814331,
      "eval_runtime": 19.752,
      "eval_samples_per_second": 59.184,
      "eval_steps_per_second": 1.873,
      "step": 6512
    },
    {
      "epoch": 45.0,
      "eval_accuracy": 0.4203195340919892,
      "eval_loss": 3.9950661659240723,
      "eval_runtime": 19.7402,
      "eval_samples_per_second": 59.219,
      "eval_steps_per_second": 1.874,
      "step": 6660
    },
    {
      "epoch": 46.0,
      "eval_accuracy": 0.42070836681615126,
      "eval_loss": 3.99338960647583,
      "eval_runtime": 19.7409,
      "eval_samples_per_second": 59.217,
      "eval_steps_per_second": 1.874,
      "step": 6808
    },
    {
      "epoch": 47.0,
      "eval_accuracy": 0.4209071035418341,
      "eval_loss": 3.990373134613037,
      "eval_runtime": 19.7427,
      "eval_samples_per_second": 59.212,
      "eval_steps_per_second": 1.874,
      "step": 6956
    },
    {
      "epoch": 47.3,
      "learning_rate": 5.405405405405406e-07,
      "loss": 3.7948,
      "step": 7000
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.4210626366314989,
      "eval_loss": 3.9910764694213867,
      "eval_runtime": 19.7469,
      "eval_samples_per_second": 59.199,
      "eval_steps_per_second": 1.874,
      "step": 7104
    },
    {
      "epoch": 49.0,
      "eval_accuracy": 0.42127001408438536,
      "eval_loss": 3.9890379905700684,
      "eval_runtime": 19.7459,
      "eval_samples_per_second": 59.202,
      "eval_steps_per_second": 1.874,
      "step": 7252
    },
    {
      "epoch": 50.0,
      "eval_accuracy": 0.42128729553879257,
      "eval_loss": 3.988962173461914,
      "eval_runtime": 19.7474,
      "eval_samples_per_second": 59.198,
      "eval_steps_per_second": 1.874,
      "step": 7400
    },
    {
      "epoch": 50.0,
      "step": 7400,
      "total_flos": 1.00265577216e+17,
      "train_loss": 4.29683648186761,
      "train_runtime": 9035.6616,
      "train_samples_per_second": 26.108,
      "train_steps_per_second": 0.819
    }
  ],
  "logging_steps": 500,
  "max_steps": 7400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "total_flos": 1.00265577216e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}