{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 40.0,
  "eval_steps": 500,
  "global_step": 10000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.4,
      "learning_rate": 0.0003,
      "loss": 1.9296,
      "step": 100
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0003,
      "loss": 1.7554,
      "step": 200
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6093164556962025,
      "eval_loss": 1.7939746379852295,
      "eval_runtime": 5.0177,
      "eval_samples_per_second": 99.648,
      "eval_steps_per_second": 12.556,
      "step": 250
    },
    {
      "epoch": 1.0,
      "eval_exact_match": 26.0,
      "eval_f1": 36.71136363636365,
      "step": 250
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0003,
      "loss": 1.6394,
      "step": 300
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0003,
      "loss": 1.5315,
      "step": 400
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0003,
      "loss": 1.5248,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6085063291139241,
      "eval_loss": 1.8273799419403076,
      "eval_runtime": 4.4023,
      "eval_samples_per_second": 113.576,
      "eval_steps_per_second": 14.311,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_exact_match": 24.4,
      "eval_f1": 33.57723554223556,
      "step": 500
    },
    {
      "epoch": 2.4,
      "learning_rate": 0.0003,
      "loss": 1.1692,
      "step": 600
    },
    {
      "epoch": 2.8,
      "learning_rate": 0.0003,
      "loss": 1.2054,
      "step": 700
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6027088607594937,
      "eval_loss": 1.9717934131622314,
      "eval_runtime": 5.1273,
      "eval_samples_per_second": 97.518,
      "eval_steps_per_second": 12.287,
      "step": 750
    },
    {
      "epoch": 3.0,
      "eval_exact_match": 20.8,
      "eval_f1": 30.584243697478996,
      "step": 750
    },
    {
      "epoch": 3.2,
      "learning_rate": 0.0003,
      "loss": 1.0252,
      "step": 800
    },
    {
      "epoch": 3.6,
      "learning_rate": 0.0003,
      "loss": 0.8608,
      "step": 900
    },
    {
      "epoch": 4.0,
      "learning_rate": 0.0003,
      "loss": 0.8989,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5986835443037974,
      "eval_loss": 2.151914596557617,
      "eval_runtime": 5.3325,
      "eval_samples_per_second": 93.765,
      "eval_steps_per_second": 11.814,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_exact_match": 21.0,
      "eval_f1": 30.363229548229558,
      "step": 1000
    },
    {
      "epoch": 4.4,
      "learning_rate": 0.0003,
      "loss": 0.5842,
      "step": 1100
    },
    {
      "epoch": 4.8,
      "learning_rate": 0.0003,
      "loss": 0.6306,
      "step": 1200
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5960506329113924,
      "eval_loss": 2.329284191131592,
      "eval_runtime": 4.5585,
      "eval_samples_per_second": 109.686,
      "eval_steps_per_second": 13.82,
      "step": 1250
    },
    {
      "epoch": 5.0,
      "eval_exact_match": 20.4,
      "eval_f1": 30.02542624042625,
      "step": 1250
    },
    {
      "epoch": 5.2,
      "learning_rate": 0.0003,
      "loss": 0.5368,
      "step": 1300
    },
    {
      "epoch": 5.6,
      "learning_rate": 0.0003,
      "loss": 0.4495,
      "step": 1400
    },
    {
      "epoch": 6.0,
      "learning_rate": 0.0003,
      "loss": 0.4712,
      "step": 1500
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5935696202531645,
      "eval_loss": 2.5598793029785156,
      "eval_runtime": 4.4094,
      "eval_samples_per_second": 113.395,
      "eval_steps_per_second": 14.288,
      "step": 1500
    },
    {
      "epoch": 6.0,
      "eval_exact_match": 20.6,
      "eval_f1": 30.108149073149082,
      "step": 1500
    },
    {
      "epoch": 6.4,
      "learning_rate": 0.0003,
      "loss": 0.3625,
      "step": 1600
    },
    {
      "epoch": 6.8,
      "learning_rate": 0.0003,
      "loss": 0.3797,
      "step": 1700
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5935696202531645,
      "eval_loss": 2.732860803604126,
      "eval_runtime": 5.2701,
      "eval_samples_per_second": 94.876,
      "eval_steps_per_second": 11.954,
      "step": 1750
    },
    {
      "epoch": 7.0,
      "eval_exact_match": 19.6,
      "eval_f1": 29.002709512709526,
      "step": 1750
    },
    {
      "epoch": 7.2,
      "learning_rate": 0.0003,
      "loss": 0.3582,
      "step": 1800
    },
    {
      "epoch": 7.6,
      "learning_rate": 0.0003,
      "loss": 0.3399,
      "step": 1900
    },
    {
      "epoch": 8.0,
      "learning_rate": 0.0003,
      "loss": 0.3527,
      "step": 2000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5912911392405064,
      "eval_loss": 2.8185083866119385,
      "eval_runtime": 4.4105,
      "eval_samples_per_second": 113.366,
      "eval_steps_per_second": 14.284,
      "step": 2000
    },
    {
      "epoch": 8.0,
      "eval_exact_match": 20.8,
      "eval_f1": 29.191525696525694,
      "step": 2000
    },
    {
      "epoch": 8.4,
      "learning_rate": 0.0003,
      "loss": 0.3202,
      "step": 2100
    },
    {
      "epoch": 8.8,
      "learning_rate": 0.0003,
      "loss": 0.3314,
      "step": 2200
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.592,
      "eval_loss": 2.824962854385376,
      "eval_runtime": 4.5596,
      "eval_samples_per_second": 109.66,
      "eval_steps_per_second": 13.817,
      "step": 2250
    },
    {
      "epoch": 9.0,
      "eval_exact_match": 21.0,
      "eval_f1": 29.434571539571547,
      "step": 2250
    },
    {
      "epoch": 9.2,
      "learning_rate": 0.0003,
      "loss": 0.3174,
      "step": 2300
    },
    {
      "epoch": 9.6,
      "learning_rate": 0.0003,
      "loss": 0.3157,
      "step": 2400
    },
    {
      "epoch": 10.0,
      "learning_rate": 0.0003,
      "loss": 0.3265,
      "step": 2500
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.5911392405063292,
      "eval_loss": 2.9242382049560547,
      "eval_runtime": 4.7857,
      "eval_samples_per_second": 104.477,
      "eval_steps_per_second": 13.164,
      "step": 2500
    },
    {
      "epoch": 10.0,
      "eval_exact_match": 20.8,
      "eval_f1": 29.524657009657016,
      "step": 2500
    },
    {
      "epoch": 10.4,
      "learning_rate": 0.0003,
      "loss": 0.2989,
      "step": 2600
    },
    {
      "epoch": 10.8,
      "learning_rate": 0.0003,
      "loss": 0.3148,
      "step": 2700
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.591240506329114,
      "eval_loss": 3.0012593269348145,
      "eval_runtime": 4.5582,
      "eval_samples_per_second": 109.691,
      "eval_steps_per_second": 13.821,
      "step": 2750
    },
    {
      "epoch": 11.0,
      "eval_exact_match": 20.0,
      "eval_f1": 28.128880727116037,
      "step": 2750
    },
    {
      "epoch": 11.2,
      "learning_rate": 0.0003,
      "loss": 0.3028,
      "step": 2800
    },
    {
      "epoch": 11.6,
      "learning_rate": 0.0003,
      "loss": 0.3047,
      "step": 2900
    },
    {
      "epoch": 12.0,
      "learning_rate": 0.0003,
      "loss": 0.3184,
      "step": 3000
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.590632911392405,
      "eval_loss": 2.931525707244873,
      "eval_runtime": 4.4989,
      "eval_samples_per_second": 111.138,
      "eval_steps_per_second": 14.003,
      "step": 3000
    },
    {
      "epoch": 12.0,
      "eval_exact_match": 21.2,
      "eval_f1": 30.12248362748364,
      "step": 3000
    },
    {
      "epoch": 12.4,
      "learning_rate": 0.0003,
      "loss": 0.3,
      "step": 3100
    },
    {
      "epoch": 12.8,
      "learning_rate": 0.0003,
      "loss": 0.3101,
      "step": 3200
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.5896962025316456,
      "eval_loss": 2.9116382598876953,
      "eval_runtime": 4.6155,
      "eval_samples_per_second": 108.331,
      "eval_steps_per_second": 13.65,
      "step": 3250
    },
    {
      "epoch": 13.0,
      "eval_exact_match": 19.2,
      "eval_f1": 27.90991785991786,
      "step": 3250
    },
    {
      "epoch": 13.2,
      "learning_rate": 0.0003,
      "loss": 0.3063,
      "step": 3300
    },
    {
      "epoch": 13.6,
      "learning_rate": 0.0003,
      "loss": 0.3041,
      "step": 3400
    },
    {
      "epoch": 14.0,
      "learning_rate": 0.0003,
      "loss": 0.3164,
      "step": 3500
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.5902025316455696,
      "eval_loss": 2.920793056488037,
      "eval_runtime": 4.4066,
      "eval_samples_per_second": 113.467,
      "eval_steps_per_second": 14.297,
      "step": 3500
    },
    {
      "epoch": 14.0,
      "eval_exact_match": 18.0,
      "eval_f1": 27.415170385170395,
      "step": 3500
    },
    {
      "epoch": 14.4,
      "learning_rate": 0.0003,
      "loss": 0.2957,
      "step": 3600
    },
    {
      "epoch": 14.8,
      "learning_rate": 0.0003,
      "loss": 0.3074,
      "step": 3700
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.5908607594936709,
      "eval_loss": 2.9385440349578857,
      "eval_runtime": 4.8175,
      "eval_samples_per_second": 103.788,
      "eval_steps_per_second": 13.077,
      "step": 3750
    },
    {
      "epoch": 15.0,
      "eval_exact_match": 21.2,
      "eval_f1": 29.687630702630706,
      "step": 3750
    },
    {
      "epoch": 15.2,
      "learning_rate": 0.0003,
      "loss": 0.3013,
      "step": 3800
    },
    {
      "epoch": 15.6,
      "learning_rate": 0.0003,
      "loss": 0.3002,
      "step": 3900
    },
    {
      "epoch": 16.0,
      "learning_rate": 0.0003,
      "loss": 0.3107,
      "step": 4000
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.5891645569620253,
      "eval_loss": 2.9519243240356445,
      "eval_runtime": 4.5619,
      "eval_samples_per_second": 109.603,
      "eval_steps_per_second": 13.81,
      "step": 4000
    },
    {
      "epoch": 16.0,
      "eval_exact_match": 19.8,
      "eval_f1": 28.670117105117104,
      "step": 4000
    },
    {
      "epoch": 16.4,
      "learning_rate": 0.0003,
      "loss": 0.2892,
      "step": 4100
    },
    {
      "epoch": 16.8,
      "learning_rate": 0.0003,
      "loss": 0.3054,
      "step": 4200
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.5898227848101266,
      "eval_loss": 3.010847568511963,
      "eval_runtime": 4.718,
      "eval_samples_per_second": 105.977,
      "eval_steps_per_second": 13.353,
      "step": 4250
    },
    {
      "epoch": 17.0,
      "eval_exact_match": 19.6,
      "eval_f1": 27.744197632432932,
      "step": 4250
    },
    {
      "epoch": 17.2,
      "learning_rate": 0.0003,
      "loss": 0.2959,
      "step": 4300
    },
    {
      "epoch": 17.6,
      "learning_rate": 0.0003,
      "loss": 0.297,
      "step": 4400
    },
    {
      "epoch": 18.0,
      "learning_rate": 0.0003,
      "loss": 0.309,
      "step": 4500
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.5903544303797469,
      "eval_loss": 3.003683567047119,
      "eval_runtime": 4.4085,
      "eval_samples_per_second": 113.418,
      "eval_steps_per_second": 14.291,
      "step": 4500
    },
    {
      "epoch": 18.0,
      "eval_exact_match": 20.6,
      "eval_f1": 28.64154694978225,
      "step": 4500
    },
    {
      "epoch": 18.4,
      "learning_rate": 0.0003,
      "loss": 0.2883,
      "step": 4600
    },
    {
      "epoch": 18.8,
      "learning_rate": 0.0003,
      "loss": 0.3005,
      "step": 4700
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.5898481012658228,
      "eval_loss": 3.0279438495635986,
      "eval_runtime": 4.7099,
      "eval_samples_per_second": 106.16,
      "eval_steps_per_second": 13.376,
      "step": 4750
    },
    {
      "epoch": 19.0,
      "eval_exact_match": 21.0,
      "eval_f1": 28.707790706026003,
      "step": 4750
    },
    {
      "epoch": 19.2,
      "learning_rate": 0.0003,
      "loss": 0.2959,
      "step": 4800
    },
    {
      "epoch": 19.6,
      "learning_rate": 0.0003,
      "loss": 0.2911,
      "step": 4900
    },
    {
      "epoch": 20.0,
      "learning_rate": 0.0003,
      "loss": 0.3127,
      "step": 5000
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.5883291139240506,
      "eval_loss": 2.9650285243988037,
      "eval_runtime": 4.711,
      "eval_samples_per_second": 106.135,
      "eval_steps_per_second": 13.373,
      "step": 5000
    },
    {
      "epoch": 20.0,
      "eval_exact_match": 11.4,
      "eval_f1": 20.32415197623728,
      "step": 5000
    },
    {
      "epoch": 20.4,
      "learning_rate": 0.0003,
      "loss": 0.2868,
      "step": 5100
    },
    {
      "epoch": 20.8,
      "learning_rate": 0.0003,
      "loss": 0.2989,
      "step": 5200
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.5895696202531645,
      "eval_loss": 3.013432502746582,
      "eval_runtime": 4.8101,
      "eval_samples_per_second": 103.947,
      "eval_steps_per_second": 13.097,
      "step": 5250
    },
    {
      "epoch": 21.0,
      "eval_exact_match": 10.8,
      "eval_f1": 17.55567536105352,
      "step": 5250
    },
    {
      "epoch": 21.2,
      "learning_rate": 0.0003,
      "loss": 0.2948,
      "step": 5300
    },
    {
      "epoch": 21.6,
      "learning_rate": 0.0003,
      "loss": 0.2921,
      "step": 5400
    },
    {
      "epoch": 22.0,
      "learning_rate": 0.0003,
      "loss": 0.3072,
      "step": 5500
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.5885822784810126,
      "eval_loss": 3.0298521518707275,
      "eval_runtime": 4.7113,
      "eval_samples_per_second": 106.128,
      "eval_steps_per_second": 13.372,
      "step": 5500
    },
    {
      "epoch": 22.0,
      "eval_exact_match": 0.0,
      "eval_f1": 0.7666666666666667,
      "step": 5500
    },
    {
      "epoch": 22.4,
      "learning_rate": 0.0003,
      "loss": 0.2914,
      "step": 5600
    },
    {
      "epoch": 22.8,
      "learning_rate": 0.0003,
      "loss": 0.2982,
      "step": 5700
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.5870126582278481,
      "eval_loss": 3.0277092456817627,
      "eval_runtime": 4.4065,
      "eval_samples_per_second": 113.468,
      "eval_steps_per_second": 14.297,
      "step": 5750
    },
    {
      "epoch": 23.0,
      "eval_exact_match": 7.6,
      "eval_f1": 9.586507936507935,
      "step": 5750
    },
    {
      "epoch": 23.2,
      "learning_rate": 0.0003,
      "loss": 0.3014,
      "step": 5800
    },
    {
      "epoch": 23.6,
      "learning_rate": 0.0003,
      "loss": 0.2907,
      "step": 5900
    },
    {
      "epoch": 24.0,
      "learning_rate": 0.0003,
      "loss": 0.3029,
      "step": 6000
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.5877215189873418,
      "eval_loss": 3.02402925491333,
      "eval_runtime": 4.7089,
      "eval_samples_per_second": 106.182,
      "eval_steps_per_second": 13.379,
      "step": 6000
    },
    {
      "epoch": 24.0,
      "eval_exact_match": 4.2,
      "eval_f1": 7.583745822888591,
      "step": 6000
    },
    {
      "epoch": 24.4,
      "learning_rate": 0.0003,
      "loss": 0.2853,
      "step": 6100
    },
    {
      "epoch": 24.8,
      "learning_rate": 0.0003,
      "loss": 0.2989,
      "step": 6200
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.5874430379746836,
      "eval_loss": 3.0006086826324463,
      "eval_runtime": 4.5617,
      "eval_samples_per_second": 109.609,
      "eval_steps_per_second": 13.811,
      "step": 6250
    },
    {
      "epoch": 25.0,
      "eval_exact_match": 0.0,
      "eval_f1": 0.5121500721500722,
      "step": 6250
    },
    {
      "epoch": 25.2,
      "learning_rate": 0.0003,
      "loss": 0.2904,
      "step": 6300
    },
    {
      "epoch": 25.6,
      "learning_rate": 0.0003,
      "loss": 0.29,
      "step": 6400
    },
    {
      "epoch": 26.0,
      "learning_rate": 0.0003,
      "loss": 0.3009,
      "step": 6500
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.5873417721518988,
      "eval_loss": 3.058345317840576,
      "eval_runtime": 4.708,
      "eval_samples_per_second": 106.202,
      "eval_steps_per_second": 13.381,
      "step": 6500
    },
    {
      "epoch": 26.0,
      "eval_exact_match": 2.2,
      "eval_f1": 4.8883016983016985,
      "step": 6500
    },
    {
      "epoch": 26.4,
      "learning_rate": 0.0003,
      "loss": 0.2793,
      "step": 6600
    },
    {
      "epoch": 26.8,
      "learning_rate": 0.0003,
      "loss": 0.2905,
      "step": 6700
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.5867848101265822,
      "eval_loss": 3.1007750034332275,
      "eval_runtime": 4.5013,
      "eval_samples_per_second": 111.079,
      "eval_steps_per_second": 13.996,
      "step": 6750
    },
    {
      "epoch": 27.0,
      "eval_exact_match": 1.2,
      "eval_f1": 2.0656084656084652,
      "step": 6750
    },
    {
      "epoch": 27.2,
      "learning_rate": 0.0003,
      "loss": 0.2964,
      "step": 6800
    },
    {
      "epoch": 27.6,
      "learning_rate": 0.0003,
      "loss": 0.3021,
      "step": 6900
    },
    {
      "epoch": 28.0,
      "learning_rate": 0.0003,
      "loss": 0.3237,
      "step": 7000
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.5851139240506329,
      "eval_loss": 3.192138671875,
      "eval_runtime": 4.5595,
      "eval_samples_per_second": 109.661,
      "eval_steps_per_second": 13.817,
      "step": 7000
    },
    {
      "epoch": 28.0,
      "eval_exact_match": 0.8,
      "eval_f1": 1.5059791467236656,
      "step": 7000
    },
    {
      "epoch": 28.4,
      "learning_rate": 0.0003,
      "loss": 0.3584,
      "step": 7100
    },
    {
      "epoch": 28.8,
      "learning_rate": 0.0003,
      "loss": 0.4738,
      "step": 7200
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.5789873417721519,
      "eval_loss": 3.32121205329895,
      "eval_runtime": 4.4073,
      "eval_samples_per_second": 113.447,
      "eval_steps_per_second": 14.294,
      "step": 7250
    },
    {
      "epoch": 29.0,
      "eval_exact_match": 3.8,
      "eval_f1": 6.689863653841047,
      "step": 7250
    },
    {
      "epoch": 29.2,
      "learning_rate": 0.0003,
      "loss": 0.8318,
      "step": 7300
    },
    {
      "epoch": 29.6,
      "learning_rate": 0.0003,
      "loss": 1.646,
      "step": 7400
    },
    {
      "epoch": 30.0,
      "learning_rate": 0.0003,
      "loss": 2.5106,
      "step": 7500
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.5371645569620254,
      "eval_loss": 3.9834704399108887,
      "eval_runtime": 5.1142,
      "eval_samples_per_second": 97.766,
      "eval_steps_per_second": 12.319,
      "step": 7500
    },
    {
      "epoch": 30.0,
      "eval_exact_match": 3.8,
      "eval_f1": 5.247779866332498,
      "step": 7500
    },
    {
      "epoch": 30.4,
      "learning_rate": 0.0003,
      "loss": 2.5833,
      "step": 7600
    },
    {
      "epoch": 30.8,
      "learning_rate": 0.0003,
      "loss": 2.4442,
      "step": 7700
    },
    {
      "epoch": 31.0,
      "eval_accuracy": 0.5435443037974683,
      "eval_loss": 3.8099365234375,
      "eval_runtime": 5.1354,
      "eval_samples_per_second": 97.364,
      "eval_steps_per_second": 12.268,
      "step": 7750
    },
    {
      "epoch": 31.0,
      "eval_exact_match": 8.0,
      "eval_f1": 11.652945717183004,
      "step": 7750
    },
    {
      "epoch": 31.2,
      "learning_rate": 0.0003,
      "loss": 2.1299,
      "step": 7800
    },
    {
      "epoch": 31.6,
      "learning_rate": 0.0003,
      "loss": 2.3675,
      "step": 7900
    },
    {
      "epoch": 32.0,
      "learning_rate": 0.0003,
      "loss": 2.5329,
      "step": 8000
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.5304810126582279,
      "eval_loss": 4.046270847320557,
      "eval_runtime": 5.1297,
      "eval_samples_per_second": 97.471,
      "eval_steps_per_second": 12.281,
      "step": 8000
    },
    {
      "epoch": 32.0,
      "eval_exact_match": 8.0,
      "eval_f1": 11.118349417066474,
      "step": 8000
    },
    {
      "epoch": 32.4,
      "learning_rate": 0.0003,
      "loss": 2.9966,
      "step": 8100
    },
    {
      "epoch": 32.8,
      "learning_rate": 0.0003,
      "loss": 3.9607,
      "step": 8200
    },
    {
      "epoch": 33.0,
      "eval_accuracy": 0.45944303797468355,
      "eval_loss": 5.853953838348389,
      "eval_runtime": 4.7634,
      "eval_samples_per_second": 104.966,
      "eval_steps_per_second": 13.226,
      "step": 8250
    },
    {
      "epoch": 33.0,
      "eval_exact_match": 0.8,
      "eval_f1": 2.1406916099773245,
      "step": 8250
    },
    {
      "epoch": 33.2,
      "learning_rate": 0.0003,
      "loss": 5.3072,
      "step": 8300
    },
    {
      "epoch": 33.6,
      "learning_rate": 0.0003,
      "loss": 8.0875,
      "step": 8400
    },
    {
      "epoch": 34.0,
      "learning_rate": 0.0003,
      "loss": 8.2643,
      "step": 8500
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.36860759493670886,
      "eval_loss": 9.332476615905762,
      "eval_runtime": 4.4983,
      "eval_samples_per_second": 111.154,
      "eval_steps_per_second": 14.005,
      "step": 8500
    },
    {
      "epoch": 34.0,
      "eval_exact_match": 0.0,
      "eval_f1": 0.12853014882332628,
      "step": 8500
    },
    {
      "epoch": 34.4,
      "learning_rate": 0.0003,
      "loss": 8.6504,
      "step": 8600
    },
    {
      "epoch": 34.8,
      "learning_rate": 0.0003,
      "loss": 8.6872,
      "step": 8700
    },
    {
      "epoch": 35.0,
      "eval_accuracy": 0.34237974683544303,
      "eval_loss": 10.414295196533203,
      "eval_runtime": 4.561,
      "eval_samples_per_second": 109.625,
      "eval_steps_per_second": 13.813,
      "step": 8750
    },
    {
      "epoch": 35.0,
      "eval_exact_match": 0.0,
      "eval_f1": 0.11605853329133448,
      "step": 8750
    },
    {
      "epoch": 35.2,
      "learning_rate": 0.0003,
      "loss": 9.9152,
      "step": 8800
    },
    {
      "epoch": 35.6,
      "learning_rate": 0.0003,
      "loss": 4.7671,
      "step": 8900
    },
    {
      "epoch": 36.0,
      "learning_rate": 0.0003,
      "loss": 0.5615,
      "step": 9000
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.5844810126582278,
      "eval_loss": 2.786999464035034,
      "eval_runtime": 5.1493,
      "eval_samples_per_second": 97.101,
      "eval_steps_per_second": 12.235,
      "step": 9000
    },
    {
      "epoch": 36.0,
      "eval_exact_match": 0.0,
      "eval_f1": 0.0,
      "step": 9000
    },
    {
      "epoch": 36.4,
      "learning_rate": 0.0003,
      "loss": 0.434,
      "step": 9100
    },
    {
      "epoch": 36.8,
      "learning_rate": 0.0003,
      "loss": 0.4658,
      "step": 9200
    },
    {
      "epoch": 37.0,
      "eval_accuracy": 0.5855189873417721,
      "eval_loss": 2.7785391807556152,
      "eval_runtime": 4.403,
      "eval_samples_per_second": 113.558,
      "eval_steps_per_second": 14.308,
      "step": 9250
    },
    {
      "epoch": 37.0,
      "eval_exact_match": 0.2,
      "eval_f1": 0.2,
      "step": 9250
    },
    {
      "epoch": 37.2,
      "learning_rate": 0.0003,
      "loss": 0.4142,
      "step": 9300
    },
    {
      "epoch": 37.6,
      "learning_rate": 0.0003,
      "loss": 0.3799,
      "step": 9400
    },
    {
      "epoch": 38.0,
      "learning_rate": 0.0003,
      "loss": 0.3872,
      "step": 9500
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 0.585873417721519,
      "eval_loss": 2.8146767616271973,
      "eval_runtime": 4.5598,
      "eval_samples_per_second": 109.653,
      "eval_steps_per_second": 13.816,
      "step": 9500
    },
    {
      "epoch": 38.0,
      "eval_exact_match": 0.0,
      "eval_f1": 0.1949989841186112,
      "step": 9500
    },
    {
      "epoch": 38.4,
      "learning_rate": 0.0003,
      "loss": 0.3113,
      "step": 9600
    },
    {
      "epoch": 38.8,
      "learning_rate": 0.0003,
      "loss": 0.3237,
      "step": 9700
    },
    {
      "epoch": 39.0,
      "eval_accuracy": 0.5856455696202532,
      "eval_loss": 2.956742286682129,
      "eval_runtime": 4.7632,
      "eval_samples_per_second": 104.972,
      "eval_steps_per_second": 13.227,
      "step": 9750
    },
    {
      "epoch": 39.0,
      "eval_exact_match": 0.2,
      "eval_f1": 0.28,
      "step": 9750
    },
    {
      "epoch": 39.2,
      "learning_rate": 0.0003,
      "loss": 0.3078,
      "step": 9800
    },
    {
      "epoch": 39.6,
      "learning_rate": 0.0003,
      "loss": 0.2905,
      "step": 9900
    },
    {
      "epoch": 40.0,
      "learning_rate": 0.0003,
      "loss": 0.3006,
      "step": 10000
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.5861518987341772,
      "eval_loss": 3.0283753871917725,
      "eval_runtime": 4.4059,
      "eval_samples_per_second": 113.485,
      "eval_steps_per_second": 14.299,
      "step": 10000
    },
    {
      "epoch": 40.0,
      "eval_exact_match": 0.0,
      "eval_f1": 0.0,
      "step": 10000
    }
  ],
  "logging_steps": 100,
  "max_steps": 12500,
  "num_train_epochs": 50,
  "save_steps": 500,
  "total_flos": 6.394690327847895e+17,
  "trial_name": null,
  "trial_params": null
}