{
  "best_metric": 0.9020043015480042,
  "best_model_checkpoint": "./output_v2/7b_cluster04_Nous-Hermes-llama-2-7b_partitioned_v3_standardized_04/checkpoint-400",
  "epoch": 0.36190907034607556,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 0.0002,
      "loss": 0.9647,
      "step": 10
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.9474,
      "step": 20
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.9287,
      "step": 30
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0002,
      "loss": 0.8723,
      "step": 40
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8812,
      "step": 50
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.8627,
      "step": 60
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.8893,
      "step": 70
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.8793,
      "step": 80
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.8778,
      "step": 90
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.8809,
      "step": 100
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.8534,
      "step": 110
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.8566,
      "step": 120
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.8979,
      "step": 130
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0002,
      "loss": 0.8919,
      "step": 140
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002,
      "loss": 0.8435,
      "step": 150
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002,
      "loss": 0.8764,
      "step": 160
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 0.8806,
      "step": 170
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0002,
      "loss": 0.8178,
      "step": 180
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002,
      "loss": 0.8821,
      "step": 190
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002,
      "loss": 0.8603,
      "step": 200
    },
    {
      "epoch": 0.18,
      "eval_loss": 0.91291344165802,
      "eval_runtime": 191.1983,
      "eval_samples_per_second": 5.23,
      "eval_steps_per_second": 2.615,
      "step": 200
    },
    {
      "epoch": 0.18,
      "mmlu_eval_accuracy": 0.4726714210605103,
      "mmlu_eval_accuracy_abstract_algebra": 0.09090909090909091,
      "mmlu_eval_accuracy_anatomy": 0.5714285714285714,
      "mmlu_eval_accuracy_astronomy": 0.4375,
      "mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
      "mmlu_eval_accuracy_clinical_knowledge": 0.4482758620689655,
      "mmlu_eval_accuracy_college_biology": 0.4375,
      "mmlu_eval_accuracy_college_chemistry": 0.125,
      "mmlu_eval_accuracy_college_computer_science": 0.45454545454545453,
      "mmlu_eval_accuracy_college_mathematics": 0.18181818181818182,
      "mmlu_eval_accuracy_college_medicine": 0.2727272727272727,
      "mmlu_eval_accuracy_college_physics": 0.45454545454545453,
      "mmlu_eval_accuracy_computer_security": 0.36363636363636365,
      "mmlu_eval_accuracy_conceptual_physics": 0.38461538461538464,
      "mmlu_eval_accuracy_econometrics": 0.16666666666666666,
      "mmlu_eval_accuracy_electrical_engineering": 0.4375,
      "mmlu_eval_accuracy_elementary_mathematics": 0.34146341463414637,
      "mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
      "mmlu_eval_accuracy_global_facts": 0.6,
      "mmlu_eval_accuracy_high_school_biology": 0.34375,
      "mmlu_eval_accuracy_high_school_chemistry": 0.4090909090909091,
      "mmlu_eval_accuracy_high_school_computer_science": 0.7777777777777778,
      "mmlu_eval_accuracy_high_school_european_history": 0.7222222222222222,
      "mmlu_eval_accuracy_high_school_geography": 0.7727272727272727,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.3488372093023256,
      "mmlu_eval_accuracy_high_school_mathematics": 0.20689655172413793,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.4230769230769231,
      "mmlu_eval_accuracy_high_school_physics": 0.35294117647058826,
      "mmlu_eval_accuracy_high_school_psychology": 0.6833333333333333,
      "mmlu_eval_accuracy_high_school_statistics": 0.34782608695652173,
      "mmlu_eval_accuracy_high_school_us_history": 0.6818181818181818,
      "mmlu_eval_accuracy_high_school_world_history": 0.5,
      "mmlu_eval_accuracy_human_aging": 0.6521739130434783,
      "mmlu_eval_accuracy_human_sexuality": 0.5,
      "mmlu_eval_accuracy_international_law": 0.6923076923076923,
      "mmlu_eval_accuracy_jurisprudence": 0.2727272727272727,
      "mmlu_eval_accuracy_logical_fallacies": 0.5555555555555556,
      "mmlu_eval_accuracy_machine_learning": 0.2727272727272727,
      "mmlu_eval_accuracy_management": 0.8181818181818182,
      "mmlu_eval_accuracy_marketing": 0.8,
      "mmlu_eval_accuracy_medical_genetics": 0.7272727272727273,
      "mmlu_eval_accuracy_miscellaneous": 0.686046511627907,
      "mmlu_eval_accuracy_moral_disputes": 0.4473684210526316,
      "mmlu_eval_accuracy_moral_scenarios": 0.24,
      "mmlu_eval_accuracy_nutrition": 0.5757575757575758,
      "mmlu_eval_accuracy_philosophy": 0.5,
      "mmlu_eval_accuracy_prehistory": 0.45714285714285713,
      "mmlu_eval_accuracy_professional_accounting": 0.25806451612903225,
      "mmlu_eval_accuracy_professional_law": 0.3588235294117647,
      "mmlu_eval_accuracy_professional_medicine": 0.3870967741935484,
      "mmlu_eval_accuracy_professional_psychology": 0.391304347826087,
      "mmlu_eval_accuracy_public_relations": 0.5,
      "mmlu_eval_accuracy_security_studies": 0.5925925925925926,
      "mmlu_eval_accuracy_sociology": 0.6363636363636364,
      "mmlu_eval_accuracy_us_foreign_policy": 0.6363636363636364,
      "mmlu_eval_accuracy_virology": 0.4444444444444444,
      "mmlu_eval_accuracy_world_religions": 0.6842105263157895,
      "mmlu_loss": 1.0958761447948202,
      "step": 200
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002,
      "loss": 0.8708,
      "step": 210
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002,
      "loss": 0.8801,
      "step": 220
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002,
      "loss": 0.8636,
      "step": 230
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 0.8371,
      "step": 240
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002,
      "loss": 0.8593,
      "step": 250
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002,
      "loss": 0.8254,
      "step": 260
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002,
      "loss": 0.8625,
      "step": 270
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002,
      "loss": 0.8968,
      "step": 280
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002,
      "loss": 0.8849,
      "step": 290
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002,
      "loss": 0.8063,
      "step": 300
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0002,
      "loss": 0.8439,
      "step": 310
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 0.8491,
      "step": 320
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002,
      "loss": 0.8443,
      "step": 330
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002,
      "loss": 0.8588,
      "step": 340
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002,
      "loss": 0.8509,
      "step": 350
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002,
      "loss": 0.8041,
      "step": 360
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002,
      "loss": 0.8418,
      "step": 370
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002,
      "loss": 0.8496,
      "step": 380
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0002,
      "loss": 0.8175,
      "step": 390
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 0.8622,
      "step": 400
    },
    {
      "epoch": 0.36,
      "eval_loss": 0.9020043015480042,
      "eval_runtime": 191.4718,
      "eval_samples_per_second": 5.223,
      "eval_steps_per_second": 2.611,
      "step": 400
    },
    {
      "epoch": 0.36,
      "mmlu_eval_accuracy": 0.4652294198230619,
      "mmlu_eval_accuracy_abstract_algebra": 0.09090909090909091,
      "mmlu_eval_accuracy_anatomy": 0.5714285714285714,
      "mmlu_eval_accuracy_astronomy": 0.4375,
      "mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
      "mmlu_eval_accuracy_clinical_knowledge": 0.41379310344827586,
      "mmlu_eval_accuracy_college_biology": 0.4375,
      "mmlu_eval_accuracy_college_chemistry": 0.125,
      "mmlu_eval_accuracy_college_computer_science": 0.36363636363636365,
      "mmlu_eval_accuracy_college_mathematics": 0.18181818181818182,
      "mmlu_eval_accuracy_college_medicine": 0.3181818181818182,
      "mmlu_eval_accuracy_college_physics": 0.45454545454545453,
      "mmlu_eval_accuracy_computer_security": 0.36363636363636365,
      "mmlu_eval_accuracy_conceptual_physics": 0.34615384615384615,
      "mmlu_eval_accuracy_econometrics": 0.16666666666666666,
      "mmlu_eval_accuracy_electrical_engineering": 0.4375,
      "mmlu_eval_accuracy_elementary_mathematics": 0.3170731707317073,
      "mmlu_eval_accuracy_formal_logic": 0.2857142857142857,
      "mmlu_eval_accuracy_global_facts": 0.6,
      "mmlu_eval_accuracy_high_school_biology": 0.40625,
      "mmlu_eval_accuracy_high_school_chemistry": 0.4090909090909091,
      "mmlu_eval_accuracy_high_school_computer_science": 0.7777777777777778,
      "mmlu_eval_accuracy_high_school_european_history": 0.5555555555555556,
      "mmlu_eval_accuracy_high_school_geography": 0.7272727272727273,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.3488372093023256,
      "mmlu_eval_accuracy_high_school_mathematics": 0.2413793103448276,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.4230769230769231,
      "mmlu_eval_accuracy_high_school_physics": 0.29411764705882354,
      "mmlu_eval_accuracy_high_school_psychology": 0.7,
      "mmlu_eval_accuracy_high_school_statistics": 0.2608695652173913,
      "mmlu_eval_accuracy_high_school_us_history": 0.6363636363636364,
      "mmlu_eval_accuracy_high_school_world_history": 0.5384615384615384,
      "mmlu_eval_accuracy_human_aging": 0.6956521739130435,
      "mmlu_eval_accuracy_human_sexuality": 0.4166666666666667,
      "mmlu_eval_accuracy_international_law": 0.7692307692307693,
      "mmlu_eval_accuracy_jurisprudence": 0.36363636363636365,
      "mmlu_eval_accuracy_logical_fallacies": 0.5555555555555556,
      "mmlu_eval_accuracy_machine_learning": 0.18181818181818182,
      "mmlu_eval_accuracy_management": 0.7272727272727273,
      "mmlu_eval_accuracy_marketing": 0.76,
      "mmlu_eval_accuracy_medical_genetics": 0.7272727272727273,
      "mmlu_eval_accuracy_miscellaneous": 0.686046511627907,
      "mmlu_eval_accuracy_moral_disputes": 0.47368421052631576,
      "mmlu_eval_accuracy_moral_scenarios": 0.24,
      "mmlu_eval_accuracy_nutrition": 0.6060606060606061,
      "mmlu_eval_accuracy_philosophy": 0.4411764705882353,
      "mmlu_eval_accuracy_prehistory": 0.42857142857142855,
      "mmlu_eval_accuracy_professional_accounting": 0.2903225806451613,
      "mmlu_eval_accuracy_professional_law": 0.3411764705882353,
      "mmlu_eval_accuracy_professional_medicine": 0.3870967741935484,
      "mmlu_eval_accuracy_professional_psychology": 0.391304347826087,
      "mmlu_eval_accuracy_public_relations": 0.5,
      "mmlu_eval_accuracy_security_studies": 0.5555555555555556,
      "mmlu_eval_accuracy_sociology": 0.6818181818181818,
      "mmlu_eval_accuracy_us_foreign_policy": 0.6363636363636364,
      "mmlu_eval_accuracy_virology": 0.4444444444444444,
      "mmlu_eval_accuracy_world_religions": 0.6842105263157895,
      "mmlu_loss": 1.1965837302975182,
      "step": 400
    }
  ],
  "max_steps": 5000,
  "num_train_epochs": 5,
  "total_flos": 9.122622440649523e+16,
  "trial_name": null,
  "trial_params": null
}