{
  "best_metric": 0.6735566258430481,
  "best_model_checkpoint": "./output_v2/7b_cluster01_Nous-Hermes-llama-2-7b_partitioned_v3_standardized_01/checkpoint-400",
  "epoch": 0.604001510003775,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 0.0002,
      "loss": 0.7673,
      "step": 10
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7735,
      "step": 20
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.7443,
      "step": 30
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0002,
      "loss": 0.6747,
      "step": 40
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.672,
      "step": 50
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.6703,
      "step": 60
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0002,
      "loss": 0.6624,
      "step": 70
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.6804,
      "step": 80
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002,
      "loss": 0.6378,
      "step": 90
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 0.6781,
      "step": 100
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002,
      "loss": 0.6059,
      "step": 110
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0002,
      "loss": 0.6616,
      "step": 120
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002,
      "loss": 0.6895,
      "step": 130
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0002,
      "loss": 0.6842,
      "step": 140
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0002,
      "loss": 0.6419,
      "step": 150
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002,
      "loss": 0.6201,
      "step": 160
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0002,
      "loss": 0.6317,
      "step": 170
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002,
      "loss": 0.6828,
      "step": 180
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 0.6586,
      "step": 190
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0002,
      "loss": 0.663,
      "step": 200
    },
    {
      "epoch": 0.3,
      "eval_loss": 0.6853392720222473,
      "eval_runtime": 230.5166,
      "eval_samples_per_second": 4.338,
      "eval_steps_per_second": 2.169,
      "step": 200
    },
    {
      "epoch": 0.3,
      "mmlu_eval_accuracy": 0.4676449470057811,
      "mmlu_eval_accuracy_abstract_algebra": 0.18181818181818182,
      "mmlu_eval_accuracy_anatomy": 0.6428571428571429,
      "mmlu_eval_accuracy_astronomy": 0.4375,
      "mmlu_eval_accuracy_business_ethics": 0.5454545454545454,
      "mmlu_eval_accuracy_clinical_knowledge": 0.4482758620689655,
      "mmlu_eval_accuracy_college_biology": 0.5,
      "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": 0.2727272727272727,
      "mmlu_eval_accuracy_conceptual_physics": 0.4230769230769231,
      "mmlu_eval_accuracy_econometrics": 0.16666666666666666,
      "mmlu_eval_accuracy_electrical_engineering": 0.375,
      "mmlu_eval_accuracy_elementary_mathematics": 0.36585365853658536,
      "mmlu_eval_accuracy_formal_logic": 0.2857142857142857,
      "mmlu_eval_accuracy_global_facts": 0.6,
      "mmlu_eval_accuracy_high_school_biology": 0.375,
      "mmlu_eval_accuracy_high_school_chemistry": 0.4090909090909091,
      "mmlu_eval_accuracy_high_school_computer_science": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_european_history": 0.5,
      "mmlu_eval_accuracy_high_school_geography": 0.7727272727272727,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.6190476190476191,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.3023255813953488,
      "mmlu_eval_accuracy_high_school_mathematics": 0.2413793103448276,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.4230769230769231,
      "mmlu_eval_accuracy_high_school_physics": 0.35294117647058826,
      "mmlu_eval_accuracy_high_school_psychology": 0.7,
      "mmlu_eval_accuracy_high_school_statistics": 0.21739130434782608,
      "mmlu_eval_accuracy_high_school_us_history": 0.7272727272727273,
      "mmlu_eval_accuracy_high_school_world_history": 0.5,
      "mmlu_eval_accuracy_human_aging": 0.6956521739130435,
      "mmlu_eval_accuracy_human_sexuality": 0.3333333333333333,
      "mmlu_eval_accuracy_international_law": 0.8461538461538461,
      "mmlu_eval_accuracy_jurisprudence": 0.45454545454545453,
      "mmlu_eval_accuracy_logical_fallacies": 0.5555555555555556,
      "mmlu_eval_accuracy_machine_learning": 0.2727272727272727,
      "mmlu_eval_accuracy_management": 0.5454545454545454,
      "mmlu_eval_accuracy_marketing": 0.8,
      "mmlu_eval_accuracy_medical_genetics": 0.7272727272727273,
      "mmlu_eval_accuracy_miscellaneous": 0.6744186046511628,
      "mmlu_eval_accuracy_moral_disputes": 0.39473684210526316,
      "mmlu_eval_accuracy_moral_scenarios": 0.24,
      "mmlu_eval_accuracy_nutrition": 0.5454545454545454,
      "mmlu_eval_accuracy_philosophy": 0.5,
      "mmlu_eval_accuracy_prehistory": 0.5142857142857142,
      "mmlu_eval_accuracy_professional_accounting": 0.2903225806451613,
      "mmlu_eval_accuracy_professional_law": 0.3588235294117647,
      "mmlu_eval_accuracy_professional_medicine": 0.41935483870967744,
      "mmlu_eval_accuracy_professional_psychology": 0.391304347826087,
      "mmlu_eval_accuracy_public_relations": 0.5833333333333334,
      "mmlu_eval_accuracy_security_studies": 0.5185185185185185,
      "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": 0.7368421052631579,
      "mmlu_loss": 1.0329186485851714,
      "step": 200
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002,
      "loss": 0.6087,
      "step": 210
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0002,
      "loss": 0.619,
      "step": 220
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0002,
      "loss": 0.6318,
      "step": 230
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 0.6463,
      "step": 240
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002,
      "loss": 0.6824,
      "step": 250
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002,
      "loss": 0.6566,
      "step": 260
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002,
      "loss": 0.6696,
      "step": 270
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0002,
      "loss": 0.5874,
      "step": 280
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002,
      "loss": 0.6725,
      "step": 290
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0002,
      "loss": 0.6457,
      "step": 300
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0002,
      "loss": 0.6532,
      "step": 310
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0002,
      "loss": 0.6033,
      "step": 320
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0002,
      "loss": 0.6371,
      "step": 330
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 0.6461,
      "step": 340
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0002,
      "loss": 0.6207,
      "step": 350
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0002,
      "loss": 0.6267,
      "step": 360
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0002,
      "loss": 0.6129,
      "step": 370
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0002,
      "loss": 0.677,
      "step": 380
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0002,
      "loss": 0.7014,
      "step": 390
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0002,
      "loss": 0.6801,
      "step": 400
    },
    {
      "epoch": 0.6,
      "eval_loss": 0.6735566258430481,
      "eval_runtime": 230.1155,
      "eval_samples_per_second": 4.346,
      "eval_steps_per_second": 2.173,
      "step": 400
    },
    {
      "epoch": 0.6,
      "mmlu_eval_accuracy": 0.4676078319072668,
      "mmlu_eval_accuracy_abstract_algebra": 0.18181818181818182,
      "mmlu_eval_accuracy_anatomy": 0.5714285714285714,
      "mmlu_eval_accuracy_astronomy": 0.375,
      "mmlu_eval_accuracy_business_ethics": 0.45454545454545453,
      "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": 0.3181818181818182,
      "mmlu_eval_accuracy_college_physics": 0.45454545454545453,
      "mmlu_eval_accuracy_computer_security": 0.2727272727272727,
      "mmlu_eval_accuracy_conceptual_physics": 0.46153846153846156,
      "mmlu_eval_accuracy_econometrics": 0.16666666666666666,
      "mmlu_eval_accuracy_electrical_engineering": 0.5,
      "mmlu_eval_accuracy_elementary_mathematics": 0.3170731707317073,
      "mmlu_eval_accuracy_formal_logic": 0.35714285714285715,
      "mmlu_eval_accuracy_global_facts": 0.5,
      "mmlu_eval_accuracy_high_school_biology": 0.375,
      "mmlu_eval_accuracy_high_school_chemistry": 0.45454545454545453,
      "mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
      "mmlu_eval_accuracy_high_school_european_history": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_geography": 0.8181818181818182,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.6666666666666666,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.3023255813953488,
      "mmlu_eval_accuracy_high_school_mathematics": 0.27586206896551724,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.46153846153846156,
      "mmlu_eval_accuracy_high_school_physics": 0.35294117647058826,
      "mmlu_eval_accuracy_high_school_psychology": 0.6833333333333333,
      "mmlu_eval_accuracy_high_school_statistics": 0.30434782608695654,
      "mmlu_eval_accuracy_high_school_us_history": 0.6818181818181818,
      "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": 0.18181818181818182,
      "mmlu_eval_accuracy_logical_fallacies": 0.5555555555555556,
      "mmlu_eval_accuracy_machine_learning": 0.18181818181818182,
      "mmlu_eval_accuracy_management": 0.6363636363636364,
      "mmlu_eval_accuracy_marketing": 0.76,
      "mmlu_eval_accuracy_medical_genetics": 0.7272727272727273,
      "mmlu_eval_accuracy_miscellaneous": 0.686046511627907,
      "mmlu_eval_accuracy_moral_disputes": 0.39473684210526316,
      "mmlu_eval_accuracy_moral_scenarios": 0.24,
      "mmlu_eval_accuracy_nutrition": 0.6363636363636364,
      "mmlu_eval_accuracy_philosophy": 0.5294117647058824,
      "mmlu_eval_accuracy_prehistory": 0.5142857142857142,
      "mmlu_eval_accuracy_professional_accounting": 0.3548387096774194,
      "mmlu_eval_accuracy_professional_law": 0.32941176470588235,
      "mmlu_eval_accuracy_professional_medicine": 0.41935483870967744,
      "mmlu_eval_accuracy_professional_psychology": 0.36231884057971014,
      "mmlu_eval_accuracy_public_relations": 0.5833333333333334,
      "mmlu_eval_accuracy_security_studies": 0.5555555555555556,
      "mmlu_eval_accuracy_sociology": 0.6818181818181818,
      "mmlu_eval_accuracy_us_foreign_policy": 0.5454545454545454,
      "mmlu_eval_accuracy_virology": 0.4444444444444444,
      "mmlu_eval_accuracy_world_religions": 0.7368421052631579,
      "mmlu_loss": 1.0410969870520013,
      "step": 400
    }
  ],
  "max_steps": 5000,
  "num_train_epochs": 8,
  "total_flos": 9.92288665729106e+16,
  "trial_name": null,
  "trial_params": null
}