{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 21.0,
  "eval_steps": 500,
  "global_step": 5250,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.4,
      "learning_rate": 5e-05,
      "loss": 2.1914,
      "step": 100
    },
    {
      "epoch": 0.8,
      "learning_rate": 5e-05,
      "loss": 1.798,
      "step": 200
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6067341772151899,
      "eval_loss": 1.8213441371917725,
      "eval_runtime": 4.8116,
      "eval_samples_per_second": 103.916,
      "eval_steps_per_second": 13.093,
      "step": 250
    },
    {
      "epoch": 1.0,
      "eval_exact_match": 25.4,
      "eval_f1": 35.50992875545509,
      "step": 250
    },
    {
      "epoch": 1.2,
      "learning_rate": 5e-05,
      "loss": 1.7534,
      "step": 300
    },
    {
      "epoch": 1.6,
      "learning_rate": 5e-05,
      "loss": 1.7163,
      "step": 400
    },
    {
      "epoch": 2.0,
      "learning_rate": 5e-05,
      "loss": 1.7,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6077215189873417,
      "eval_loss": 1.8046445846557617,
      "eval_runtime": 5.1228,
      "eval_samples_per_second": 97.603,
      "eval_steps_per_second": 12.298,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_exact_match": 25.8,
      "eval_f1": 35.76454372424963,
      "step": 500
    },
    {
      "epoch": 2.4,
      "learning_rate": 5e-05,
      "loss": 1.5854,
      "step": 600
    },
    {
      "epoch": 2.8,
      "learning_rate": 5e-05,
      "loss": 1.5869,
      "step": 700
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6071392405063291,
      "eval_loss": 1.829284906387329,
      "eval_runtime": 4.5692,
      "eval_samples_per_second": 109.429,
      "eval_steps_per_second": 13.788,
      "step": 750
    },
    {
      "epoch": 3.0,
      "eval_exact_match": 24.4,
      "eval_f1": 34.46831890331891,
      "step": 750
    },
    {
      "epoch": 3.2,
      "learning_rate": 5e-05,
      "loss": 1.5039,
      "step": 800
    },
    {
      "epoch": 3.6,
      "learning_rate": 5e-05,
      "loss": 1.4165,
      "step": 900
    },
    {
      "epoch": 4.0,
      "learning_rate": 5e-05,
      "loss": 1.4349,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6042784810126582,
      "eval_loss": 1.89736008644104,
      "eval_runtime": 4.5698,
      "eval_samples_per_second": 109.415,
      "eval_steps_per_second": 13.786,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_exact_match": 22.8,
      "eval_f1": 32.991320346320364,
      "step": 1000
    },
    {
      "epoch": 4.4,
      "learning_rate": 5e-05,
      "loss": 1.2624,
      "step": 1100
    },
    {
      "epoch": 4.8,
      "learning_rate": 5e-05,
      "loss": 1.3111,
      "step": 1200
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.601493670886076,
      "eval_loss": 1.9769378900527954,
      "eval_runtime": 4.7168,
      "eval_samples_per_second": 106.005,
      "eval_steps_per_second": 13.357,
      "step": 1250
    },
    {
      "epoch": 5.0,
      "eval_exact_match": 22.6,
      "eval_f1": 32.46891774891776,
      "step": 1250
    },
    {
      "epoch": 5.2,
      "learning_rate": 5e-05,
      "loss": 1.226,
      "step": 1300
    },
    {
      "epoch": 5.6,
      "learning_rate": 5e-05,
      "loss": 1.1608,
      "step": 1400
    },
    {
      "epoch": 6.0,
      "learning_rate": 5e-05,
      "loss": 1.197,
      "step": 1500
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.599240506329114,
      "eval_loss": 2.0634872913360596,
      "eval_runtime": 4.4047,
      "eval_samples_per_second": 113.515,
      "eval_steps_per_second": 14.303,
      "step": 1500
    },
    {
      "epoch": 6.0,
      "eval_exact_match": 21.4,
      "eval_f1": 29.995800865800874,
      "step": 1500
    },
    {
      "epoch": 6.4,
      "learning_rate": 5e-05,
      "loss": 1.0336,
      "step": 1600
    },
    {
      "epoch": 6.8,
      "learning_rate": 5e-05,
      "loss": 1.0729,
      "step": 1700
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5975443037974684,
      "eval_loss": 2.1523237228393555,
      "eval_runtime": 4.4025,
      "eval_samples_per_second": 113.571,
      "eval_steps_per_second": 14.31,
      "step": 1750
    },
    {
      "epoch": 7.0,
      "eval_exact_match": 20.8,
      "eval_f1": 29.054994058229354,
      "step": 1750
    },
    {
      "epoch": 7.2,
      "learning_rate": 5e-05,
      "loss": 1.0005,
      "step": 1800
    },
    {
      "epoch": 7.6,
      "learning_rate": 5e-05,
      "loss": 0.9437,
      "step": 1900
    },
    {
      "epoch": 8.0,
      "learning_rate": 5e-05,
      "loss": 0.9833,
      "step": 2000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5947341772151898,
      "eval_loss": 2.2639756202697754,
      "eval_runtime": 4.7031,
      "eval_samples_per_second": 106.312,
      "eval_steps_per_second": 13.395,
      "step": 2000
    },
    {
      "epoch": 8.0,
      "eval_exact_match": 20.8,
      "eval_f1": 28.71228732052262,
      "step": 2000
    },
    {
      "epoch": 8.4,
      "learning_rate": 5e-05,
      "loss": 0.823,
      "step": 2100
    },
    {
      "epoch": 8.8,
      "learning_rate": 5e-05,
      "loss": 0.8672,
      "step": 2200
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.5924050632911393,
      "eval_loss": 2.3642578125,
      "eval_runtime": 4.4051,
      "eval_samples_per_second": 113.505,
      "eval_steps_per_second": 14.302,
      "step": 2250
    },
    {
      "epoch": 9.0,
      "eval_exact_match": 22.4,
      "eval_f1": 30.14020120686362,
      "step": 2250
    },
    {
      "epoch": 9.2,
      "learning_rate": 5e-05,
      "loss": 0.8032,
      "step": 2300
    },
    {
      "epoch": 9.6,
      "learning_rate": 5e-05,
      "loss": 0.7566,
      "step": 2400
    },
    {
      "epoch": 10.0,
      "learning_rate": 5e-05,
      "loss": 0.7883,
      "step": 2500
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.5908101265822785,
      "eval_loss": 2.4598007202148438,
      "eval_runtime": 4.5544,
      "eval_samples_per_second": 109.783,
      "eval_steps_per_second": 13.833,
      "step": 2500
    },
    {
      "epoch": 10.0,
      "eval_exact_match": 21.4,
      "eval_f1": 29.271013039248334,
      "step": 2500
    },
    {
      "epoch": 10.4,
      "learning_rate": 5e-05,
      "loss": 0.6488,
      "step": 2600
    },
    {
      "epoch": 10.8,
      "learning_rate": 5e-05,
      "loss": 0.6879,
      "step": 2700
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.588987341772152,
      "eval_loss": 2.5668857097625732,
      "eval_runtime": 4.6769,
      "eval_samples_per_second": 106.908,
      "eval_steps_per_second": 13.47,
      "step": 2750
    },
    {
      "epoch": 11.0,
      "eval_exact_match": 20.2,
      "eval_f1": 29.08202869026399,
      "step": 2750
    },
    {
      "epoch": 11.2,
      "learning_rate": 5e-05,
      "loss": 0.6502,
      "step": 2800
    },
    {
      "epoch": 11.6,
      "learning_rate": 5e-05,
      "loss": 0.602,
      "step": 2900
    },
    {
      "epoch": 12.0,
      "learning_rate": 5e-05,
      "loss": 0.6295,
      "step": 3000
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.5885316455696202,
      "eval_loss": 2.700011730194092,
      "eval_runtime": 5.1309,
      "eval_samples_per_second": 97.45,
      "eval_steps_per_second": 12.279,
      "step": 3000
    },
    {
      "epoch": 12.0,
      "eval_exact_match": 19.6,
      "eval_f1": 27.678461864932462,
      "step": 3000
    },
    {
      "epoch": 12.4,
      "learning_rate": 5e-05,
      "loss": 0.5228,
      "step": 3100
    },
    {
      "epoch": 12.8,
      "learning_rate": 5e-05,
      "loss": 0.5545,
      "step": 3200
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.5850886075949367,
      "eval_loss": 2.8281185626983643,
      "eval_runtime": 4.5558,
      "eval_samples_per_second": 109.749,
      "eval_steps_per_second": 13.828,
      "step": 3250
    },
    {
      "epoch": 13.0,
      "eval_exact_match": 20.0,
      "eval_f1": 28.273127941889552,
      "step": 3250
    },
    {
      "epoch": 13.2,
      "learning_rate": 5e-05,
      "loss": 0.5244,
      "step": 3300
    },
    {
      "epoch": 13.6,
      "learning_rate": 5e-05,
      "loss": 0.4924,
      "step": 3400
    },
    {
      "epoch": 14.0,
      "learning_rate": 5e-05,
      "loss": 0.5208,
      "step": 3500
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.5852911392405064,
      "eval_loss": 2.879397392272949,
      "eval_runtime": 5.1301,
      "eval_samples_per_second": 97.464,
      "eval_steps_per_second": 12.28,
      "step": 3500
    },
    {
      "epoch": 14.0,
      "eval_exact_match": 20.2,
      "eval_f1": 28.16958851605911,
      "step": 3500
    },
    {
      "epoch": 14.4,
      "learning_rate": 5e-05,
      "loss": 0.4405,
      "step": 3600
    },
    {
      "epoch": 14.8,
      "learning_rate": 5e-05,
      "loss": 0.4679,
      "step": 3700
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.586253164556962,
      "eval_loss": 2.9183950424194336,
      "eval_runtime": 4.8449,
      "eval_samples_per_second": 103.201,
      "eval_steps_per_second": 13.003,
      "step": 3750
    },
    {
      "epoch": 15.0,
      "eval_exact_match": 19.8,
      "eval_f1": 27.632893381128678,
      "step": 3750
    },
    {
      "epoch": 15.2,
      "learning_rate": 5e-05,
      "loss": 0.4443,
      "step": 3800
    },
    {
      "epoch": 15.6,
      "learning_rate": 5e-05,
      "loss": 0.4235,
      "step": 3900
    },
    {
      "epoch": 16.0,
      "learning_rate": 5e-05,
      "loss": 0.4464,
      "step": 4000
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.5852151898734177,
      "eval_loss": 3.0790698528289795,
      "eval_runtime": 4.7018,
      "eval_samples_per_second": 106.343,
      "eval_steps_per_second": 13.399,
      "step": 4000
    },
    {
      "epoch": 16.0,
      "eval_exact_match": 21.0,
      "eval_f1": 29.560525356995953,
      "step": 4000
    },
    {
      "epoch": 16.4,
      "learning_rate": 5e-05,
      "loss": 0.3919,
      "step": 4100
    },
    {
      "epoch": 16.8,
      "learning_rate": 5e-05,
      "loss": 0.4136,
      "step": 4200
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.5855696202531645,
      "eval_loss": 3.0832109451293945,
      "eval_runtime": 4.7575,
      "eval_samples_per_second": 105.096,
      "eval_steps_per_second": 13.242,
      "step": 4250
    },
    {
      "epoch": 17.0,
      "eval_exact_match": 20.4,
      "eval_f1": 28.69772782772784,
      "step": 4250
    },
    {
      "epoch": 17.2,
      "learning_rate": 5e-05,
      "loss": 0.3902,
      "step": 4300
    },
    {
      "epoch": 17.6,
      "learning_rate": 5e-05,
      "loss": 0.3841,
      "step": 4400
    },
    {
      "epoch": 18.0,
      "learning_rate": 5e-05,
      "loss": 0.4021,
      "step": 4500
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.5846835443037974,
      "eval_loss": 3.0944228172302246,
      "eval_runtime": 4.7087,
      "eval_samples_per_second": 106.186,
      "eval_steps_per_second": 13.379,
      "step": 4500
    },
    {
      "epoch": 18.0,
      "eval_exact_match": 20.0,
      "eval_f1": 28.558940281940288,
      "step": 4500
    },
    {
      "epoch": 18.4,
      "learning_rate": 5e-05,
      "loss": 0.3577,
      "step": 4600
    },
    {
      "epoch": 18.8,
      "learning_rate": 5e-05,
      "loss": 0.3776,
      "step": 4700
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.5828354430379746,
      "eval_loss": 3.2120306491851807,
      "eval_runtime": 4.5614,
      "eval_samples_per_second": 109.616,
      "eval_steps_per_second": 13.812,
      "step": 4750
    },
    {
      "epoch": 19.0,
      "eval_exact_match": 20.2,
      "eval_f1": 28.91690789886443,
      "step": 4750
    },
    {
      "epoch": 19.2,
      "learning_rate": 5e-05,
      "loss": 0.3617,
      "step": 4800
    },
    {
      "epoch": 19.6,
      "learning_rate": 5e-05,
      "loss": 0.3575,
      "step": 4900
    },
    {
      "epoch": 20.0,
      "learning_rate": 5e-05,
      "loss": 0.373,
      "step": 5000
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.5839240506329114,
      "eval_loss": 3.229841947555542,
      "eval_runtime": 4.5542,
      "eval_samples_per_second": 109.788,
      "eval_steps_per_second": 13.833,
      "step": 5000
    },
    {
      "epoch": 20.0,
      "eval_exact_match": 20.0,
      "eval_f1": 29.50454927425517,
      "step": 5000
    },
    {
      "epoch": 20.4,
      "learning_rate": 5e-05,
      "loss": 0.3333,
      "step": 5100
    },
    {
      "epoch": 20.8,
      "learning_rate": 5e-05,
      "loss": 0.3572,
      "step": 5200
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.584126582278481,
      "eval_loss": 3.2433719635009766,
      "eval_runtime": 5.128,
      "eval_samples_per_second": 97.503,
      "eval_steps_per_second": 12.285,
      "step": 5250
    },
    {
      "epoch": 21.0,
      "eval_exact_match": 20.0,
      "eval_f1": 28.610755910755913,
      "step": 5250
    }
  ],
  "logging_steps": 100,
  "max_steps": 12500,
  "num_train_epochs": 50,
  "save_steps": 500,
  "total_flos": 3.3566494603555635e+17,
  "trial_name": null,
  "trial_params": null
}