{
  "best_metric": 0.4632822871208191,
  "best_model_checkpoint": "./output_v2/7b_cluster019_Nous-Hermes-llama-2-7b_partitioned_v3_standardized_019/checkpoint-600",
  "epoch": 2.038216560509554,
  "global_step": 1200,
  "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": 0.6267,
      "step": 10
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 0.7811,
      "step": 20
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0002,
      "loss": 0.5062,
      "step": 30
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 0.6137,
      "step": 40
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0002,
      "loss": 0.4957,
      "step": 50
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0002,
      "loss": 0.4838,
      "step": 60
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0002,
      "loss": 0.6938,
      "step": 70
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0002,
      "loss": 0.4848,
      "step": 80
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 0.4587,
      "step": 90
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0002,
      "loss": 0.5768,
      "step": 100
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0002,
      "loss": 0.4725,
      "step": 110
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0002,
      "loss": 0.5152,
      "step": 120
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 0.5707,
      "step": 130
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0002,
      "loss": 0.5002,
      "step": 140
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002,
      "loss": 0.4043,
      "step": 150
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002,
      "loss": 0.6542,
      "step": 160
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 0.4533,
      "step": 170
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0002,
      "loss": 0.5814,
      "step": 180
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0002,
      "loss": 0.525,
      "step": 190
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0002,
      "loss": 0.5448,
      "step": 200
    },
    {
      "epoch": 0.34,
      "eval_loss": 0.48780253529548645,
      "eval_runtime": 101.8557,
      "eval_samples_per_second": 9.818,
      "eval_steps_per_second": 4.909,
      "step": 200
    },
    {
      "epoch": 0.34,
      "mmlu_eval_accuracy": 0.4580124869645426,
      "mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
      "mmlu_eval_accuracy_anatomy": 0.6428571428571429,
      "mmlu_eval_accuracy_astronomy": 0.5,
      "mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
      "mmlu_eval_accuracy_clinical_knowledge": 0.41379310344827586,
      "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": 0.2727272727272727,
      "mmlu_eval_accuracy_college_physics": 0.5454545454545454,
      "mmlu_eval_accuracy_computer_security": 0.2727272727272727,
      "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": 0.4146341463414634,
      "mmlu_eval_accuracy_formal_logic": 0.2857142857142857,
      "mmlu_eval_accuracy_global_facts": 0.4,
      "mmlu_eval_accuracy_high_school_biology": 0.375,
      "mmlu_eval_accuracy_high_school_chemistry": 0.4090909090909091,
      "mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
      "mmlu_eval_accuracy_high_school_european_history": 0.5555555555555556,
      "mmlu_eval_accuracy_high_school_geography": 0.6363636363636364,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.5714285714285714,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.3488372093023256,
      "mmlu_eval_accuracy_high_school_mathematics": 0.27586206896551724,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.4230769230769231,
      "mmlu_eval_accuracy_high_school_physics": 0.29411764705882354,
      "mmlu_eval_accuracy_high_school_psychology": 0.7166666666666667,
      "mmlu_eval_accuracy_high_school_statistics": 0.30434782608695654,
      "mmlu_eval_accuracy_high_school_us_history": 0.6818181818181818,
      "mmlu_eval_accuracy_high_school_world_history": 0.46153846153846156,
      "mmlu_eval_accuracy_human_aging": 0.7391304347826086,
      "mmlu_eval_accuracy_human_sexuality": 0.5,
      "mmlu_eval_accuracy_international_law": 0.7692307692307693,
      "mmlu_eval_accuracy_jurisprudence": 0.2727272727272727,
      "mmlu_eval_accuracy_logical_fallacies": 0.5555555555555556,
      "mmlu_eval_accuracy_machine_learning": 0.2727272727272727,
      "mmlu_eval_accuracy_management": 0.36363636363636365,
      "mmlu_eval_accuracy_marketing": 0.68,
      "mmlu_eval_accuracy_medical_genetics": 0.7272727272727273,
      "mmlu_eval_accuracy_miscellaneous": 0.6627906976744186,
      "mmlu_eval_accuracy_moral_disputes": 0.47368421052631576,
      "mmlu_eval_accuracy_moral_scenarios": 0.23,
      "mmlu_eval_accuracy_nutrition": 0.5151515151515151,
      "mmlu_eval_accuracy_philosophy": 0.5294117647058824,
      "mmlu_eval_accuracy_prehistory": 0.5428571428571428,
      "mmlu_eval_accuracy_professional_accounting": 0.25806451612903225,
      "mmlu_eval_accuracy_professional_law": 0.3176470588235294,
      "mmlu_eval_accuracy_professional_medicine": 0.41935483870967744,
      "mmlu_eval_accuracy_professional_psychology": 0.42028985507246375,
      "mmlu_eval_accuracy_public_relations": 0.6666666666666666,
      "mmlu_eval_accuracy_security_studies": 0.4444444444444444,
      "mmlu_eval_accuracy_sociology": 0.5,
      "mmlu_eval_accuracy_us_foreign_policy": 0.6363636363636364,
      "mmlu_eval_accuracy_virology": 0.5,
      "mmlu_eval_accuracy_world_religions": 0.6842105263157895,
      "mmlu_loss": 0.9094281114112615,
      "step": 200
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 0.4536,
      "step": 210
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0002,
      "loss": 0.5147,
      "step": 220
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0002,
      "loss": 0.423,
      "step": 230
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0002,
      "loss": 0.5832,
      "step": 240
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0002,
      "loss": 0.4719,
      "step": 250
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0002,
      "loss": 0.452,
      "step": 260
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.0002,
      "loss": 0.4907,
      "step": 270
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0002,
      "loss": 0.5322,
      "step": 280
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0002,
      "loss": 0.592,
      "step": 290
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 0.5964,
      "step": 300
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0002,
      "loss": 0.5404,
      "step": 310
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0002,
      "loss": 0.5788,
      "step": 320
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0002,
      "loss": 0.4701,
      "step": 330
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0002,
      "loss": 0.4899,
      "step": 340
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0002,
      "loss": 0.5177,
      "step": 350
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0002,
      "loss": 0.479,
      "step": 360
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0002,
      "loss": 0.4815,
      "step": 370
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0002,
      "loss": 0.4935,
      "step": 380
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0002,
      "loss": 0.5712,
      "step": 390
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0002,
      "loss": 0.4873,
      "step": 400
    },
    {
      "epoch": 0.68,
      "eval_loss": 0.4672442674636841,
      "eval_runtime": 102.1346,
      "eval_samples_per_second": 9.791,
      "eval_steps_per_second": 4.896,
      "step": 400
    },
    {
      "epoch": 0.68,
      "mmlu_eval_accuracy": 0.4486531670462866,
      "mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
      "mmlu_eval_accuracy_anatomy": 0.5714285714285714,
      "mmlu_eval_accuracy_astronomy": 0.5,
      "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": 0.36363636363636365,
      "mmlu_eval_accuracy_college_mathematics": 0.2727272727272727,
      "mmlu_eval_accuracy_college_medicine": 0.2727272727272727,
      "mmlu_eval_accuracy_college_physics": 0.45454545454545453,
      "mmlu_eval_accuracy_computer_security": 0.2727272727272727,
      "mmlu_eval_accuracy_conceptual_physics": 0.38461538461538464,
      "mmlu_eval_accuracy_econometrics": 0.25,
      "mmlu_eval_accuracy_electrical_engineering": 0.3125,
      "mmlu_eval_accuracy_elementary_mathematics": 0.3902439024390244,
      "mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
      "mmlu_eval_accuracy_global_facts": 0.5,
      "mmlu_eval_accuracy_high_school_biology": 0.40625,
      "mmlu_eval_accuracy_high_school_chemistry": 0.2727272727272727,
      "mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
      "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": 0.3488372093023256,
      "mmlu_eval_accuracy_high_school_mathematics": 0.2413793103448276,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.38461538461538464,
      "mmlu_eval_accuracy_high_school_physics": 0.29411764705882354,
      "mmlu_eval_accuracy_high_school_psychology": 0.75,
      "mmlu_eval_accuracy_high_school_statistics": 0.2608695652173913,
      "mmlu_eval_accuracy_high_school_us_history": 0.6363636363636364,
      "mmlu_eval_accuracy_high_school_world_history": 0.46153846153846156,
      "mmlu_eval_accuracy_human_aging": 0.6956521739130435,
      "mmlu_eval_accuracy_human_sexuality": 0.3333333333333333,
      "mmlu_eval_accuracy_international_law": 0.7692307692307693,
      "mmlu_eval_accuracy_jurisprudence": 0.2727272727272727,
      "mmlu_eval_accuracy_logical_fallacies": 0.5555555555555556,
      "mmlu_eval_accuracy_machine_learning": 0.2727272727272727,
      "mmlu_eval_accuracy_management": 0.45454545454545453,
      "mmlu_eval_accuracy_marketing": 0.68,
      "mmlu_eval_accuracy_medical_genetics": 0.6363636363636364,
      "mmlu_eval_accuracy_miscellaneous": 0.6744186046511628,
      "mmlu_eval_accuracy_moral_disputes": 0.5,
      "mmlu_eval_accuracy_moral_scenarios": 0.25,
      "mmlu_eval_accuracy_nutrition": 0.5757575757575758,
      "mmlu_eval_accuracy_philosophy": 0.4411764705882353,
      "mmlu_eval_accuracy_prehistory": 0.5714285714285714,
      "mmlu_eval_accuracy_professional_accounting": 0.25806451612903225,
      "mmlu_eval_accuracy_professional_law": 0.32941176470588235,
      "mmlu_eval_accuracy_professional_medicine": 0.45161290322580644,
      "mmlu_eval_accuracy_professional_psychology": 0.4057971014492754,
      "mmlu_eval_accuracy_public_relations": 0.5,
      "mmlu_eval_accuracy_security_studies": 0.4444444444444444,
      "mmlu_eval_accuracy_sociology": 0.5909090909090909,
      "mmlu_eval_accuracy_us_foreign_policy": 0.7272727272727273,
      "mmlu_eval_accuracy_virology": 0.4444444444444444,
      "mmlu_eval_accuracy_world_religions": 0.7368421052631579,
      "mmlu_loss": 0.9122924784456159,
      "step": 400
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0002,
      "loss": 0.5392,
      "step": 410
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0002,
      "loss": 0.4237,
      "step": 420
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0002,
      "loss": 0.4864,
      "step": 430
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0002,
      "loss": 0.4317,
      "step": 440
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0002,
      "loss": 0.4613,
      "step": 450
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0002,
      "loss": 0.4595,
      "step": 460
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0002,
      "loss": 0.623,
      "step": 470
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0002,
      "loss": 0.5262,
      "step": 480
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.0002,
      "loss": 0.4351,
      "step": 490
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0002,
      "loss": 0.5168,
      "step": 500
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.0002,
      "loss": 0.4274,
      "step": 510
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.0002,
      "loss": 0.5015,
      "step": 520
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0002,
      "loss": 0.4768,
      "step": 530
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.0002,
      "loss": 0.4208,
      "step": 540
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0002,
      "loss": 0.4848,
      "step": 550
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0002,
      "loss": 0.4043,
      "step": 560
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0002,
      "loss": 0.4383,
      "step": 570
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0002,
      "loss": 0.5794,
      "step": 580
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0002,
      "loss": 0.439,
      "step": 590
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.0002,
      "loss": 0.3456,
      "step": 600
    },
    {
      "epoch": 1.02,
      "eval_loss": 0.4632822871208191,
      "eval_runtime": 101.8929,
      "eval_samples_per_second": 9.814,
      "eval_steps_per_second": 4.907,
      "step": 600
    },
    {
      "epoch": 1.02,
      "mmlu_eval_accuracy": 0.4617338416384464,
      "mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
      "mmlu_eval_accuracy_anatomy": 0.6428571428571429,
      "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": 0.5,
      "mmlu_eval_accuracy_college_chemistry": 0.0,
      "mmlu_eval_accuracy_college_computer_science": 0.36363636363636365,
      "mmlu_eval_accuracy_college_mathematics": 0.18181818181818182,
      "mmlu_eval_accuracy_college_medicine": 0.36363636363636365,
      "mmlu_eval_accuracy_college_physics": 0.45454545454545453,
      "mmlu_eval_accuracy_computer_security": 0.36363636363636365,
      "mmlu_eval_accuracy_conceptual_physics": 0.4230769230769231,
      "mmlu_eval_accuracy_econometrics": 0.25,
      "mmlu_eval_accuracy_electrical_engineering": 0.3125,
      "mmlu_eval_accuracy_elementary_mathematics": 0.3902439024390244,
      "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": 0.45454545454545453,
      "mmlu_eval_accuracy_high_school_computer_science": 0.5555555555555556,
      "mmlu_eval_accuracy_high_school_european_history": 0.5,
      "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": 0.1724137931034483,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.4230769230769231,
      "mmlu_eval_accuracy_high_school_physics": 0.29411764705882354,
      "mmlu_eval_accuracy_high_school_psychology": 0.7166666666666667,
      "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": 0.6521739130434783,
      "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": 0.36363636363636365,
      "mmlu_eval_accuracy_management": 0.6363636363636364,
      "mmlu_eval_accuracy_marketing": 0.68,
      "mmlu_eval_accuracy_medical_genetics": 0.7272727272727273,
      "mmlu_eval_accuracy_miscellaneous": 0.6744186046511628,
      "mmlu_eval_accuracy_moral_disputes": 0.47368421052631576,
      "mmlu_eval_accuracy_moral_scenarios": 0.26,
      "mmlu_eval_accuracy_nutrition": 0.5454545454545454,
      "mmlu_eval_accuracy_philosophy": 0.5294117647058824,
      "mmlu_eval_accuracy_prehistory": 0.4857142857142857,
      "mmlu_eval_accuracy_professional_accounting": 0.2903225806451613,
      "mmlu_eval_accuracy_professional_law": 0.35294117647058826,
      "mmlu_eval_accuracy_professional_medicine": 0.41935483870967744,
      "mmlu_eval_accuracy_professional_psychology": 0.4057971014492754,
      "mmlu_eval_accuracy_public_relations": 0.5,
      "mmlu_eval_accuracy_security_studies": 0.4074074074074074,
      "mmlu_eval_accuracy_sociology": 0.5454545454545454,
      "mmlu_eval_accuracy_us_foreign_policy": 0.5454545454545454,
      "mmlu_eval_accuracy_virology": 0.3888888888888889,
      "mmlu_eval_accuracy_world_religions": 0.7368421052631579,
      "mmlu_loss": 0.8381480119110866,
      "step": 600
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0002,
      "loss": 0.3464,
      "step": 610
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.0002,
      "loss": 0.4158,
      "step": 620
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0002,
      "loss": 0.3465,
      "step": 630
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0002,
      "loss": 0.3078,
      "step": 640
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0002,
      "loss": 0.4329,
      "step": 650
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0002,
      "loss": 0.3874,
      "step": 660
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0002,
      "loss": 0.4908,
      "step": 670
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0002,
      "loss": 0.5097,
      "step": 680
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0002,
      "loss": 0.3967,
      "step": 690
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.0002,
      "loss": 0.4721,
      "step": 700
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0002,
      "loss": 0.3612,
      "step": 710
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0002,
      "loss": 0.4453,
      "step": 720
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0002,
      "loss": 0.4538,
      "step": 730
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0002,
      "loss": 0.3903,
      "step": 740
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0002,
      "loss": 0.3541,
      "step": 750
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0002,
      "loss": 0.3564,
      "step": 760
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0002,
      "loss": 0.386,
      "step": 770
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0002,
      "loss": 0.4495,
      "step": 780
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0002,
      "loss": 0.3281,
      "step": 790
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0002,
      "loss": 0.3315,
      "step": 800
    },
    {
      "epoch": 1.36,
      "eval_loss": 0.47132888436317444,
      "eval_runtime": 102.2178,
      "eval_samples_per_second": 9.783,
      "eval_steps_per_second": 4.892,
      "step": 800
    },
    {
      "epoch": 1.36,
      "mmlu_eval_accuracy": 0.4676211978570877,
      "mmlu_eval_accuracy_abstract_algebra": 0.36363636363636365,
      "mmlu_eval_accuracy_anatomy": 0.5714285714285714,
      "mmlu_eval_accuracy_astronomy": 0.4375,
      "mmlu_eval_accuracy_business_ethics": 0.6363636363636364,
      "mmlu_eval_accuracy_clinical_knowledge": 0.4827586206896552,
      "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": 0.2727272727272727,
      "mmlu_eval_accuracy_college_medicine": 0.3181818181818182,
      "mmlu_eval_accuracy_college_physics": 0.45454545454545453,
      "mmlu_eval_accuracy_computer_security": 0.45454545454545453,
      "mmlu_eval_accuracy_conceptual_physics": 0.38461538461538464,
      "mmlu_eval_accuracy_econometrics": 0.25,
      "mmlu_eval_accuracy_electrical_engineering": 0.3125,
      "mmlu_eval_accuracy_elementary_mathematics": 0.3170731707317073,
      "mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
      "mmlu_eval_accuracy_global_facts": 0.6,
      "mmlu_eval_accuracy_high_school_biology": 0.4375,
      "mmlu_eval_accuracy_high_school_chemistry": 0.36363636363636365,
      "mmlu_eval_accuracy_high_school_computer_science": 0.4444444444444444,
      "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": 0.32558139534883723,
      "mmlu_eval_accuracy_high_school_mathematics": 0.1724137931034483,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.4230769230769231,
      "mmlu_eval_accuracy_high_school_physics": 0.35294117647058826,
      "mmlu_eval_accuracy_high_school_psychology": 0.75,
      "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": 0.7391304347826086,
      "mmlu_eval_accuracy_human_sexuality": 0.3333333333333333,
      "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": 0.36363636363636365,
      "mmlu_eval_accuracy_management": 0.5454545454545454,
      "mmlu_eval_accuracy_marketing": 0.68,
      "mmlu_eval_accuracy_medical_genetics": 0.7272727272727273,
      "mmlu_eval_accuracy_miscellaneous": 0.6744186046511628,
      "mmlu_eval_accuracy_moral_disputes": 0.5,
      "mmlu_eval_accuracy_moral_scenarios": 0.29,
      "mmlu_eval_accuracy_nutrition": 0.5757575757575758,
      "mmlu_eval_accuracy_philosophy": 0.5,
      "mmlu_eval_accuracy_prehistory": 0.5142857142857142,
      "mmlu_eval_accuracy_professional_accounting": 0.2903225806451613,
      "mmlu_eval_accuracy_professional_law": 0.34705882352941175,
      "mmlu_eval_accuracy_professional_medicine": 0.3870967741935484,
      "mmlu_eval_accuracy_professional_psychology": 0.4492753623188406,
      "mmlu_eval_accuracy_public_relations": 0.6666666666666666,
      "mmlu_eval_accuracy_security_studies": 0.37037037037037035,
      "mmlu_eval_accuracy_sociology": 0.5909090909090909,
      "mmlu_eval_accuracy_us_foreign_policy": 0.7272727272727273,
      "mmlu_eval_accuracy_virology": 0.3888888888888889,
      "mmlu_eval_accuracy_world_religions": 0.7368421052631579,
      "mmlu_loss": 0.8817933564495013,
      "step": 800
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0002,
      "loss": 0.4534,
      "step": 810
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0002,
      "loss": 0.3944,
      "step": 820
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0002,
      "loss": 0.4096,
      "step": 830
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.0002,
      "loss": 0.4039,
      "step": 840
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0002,
      "loss": 0.438,
      "step": 850
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0002,
      "loss": 0.3773,
      "step": 860
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0002,
      "loss": 0.4969,
      "step": 870
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0002,
      "loss": 0.396,
      "step": 880
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.0002,
      "loss": 0.4196,
      "step": 890
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0002,
      "loss": 0.5202,
      "step": 900
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0002,
      "loss": 0.4728,
      "step": 910
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0002,
      "loss": 0.4229,
      "step": 920
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.0002,
      "loss": 0.4879,
      "step": 930
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0002,
      "loss": 0.4288,
      "step": 940
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0002,
      "loss": 0.4085,
      "step": 950
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0002,
      "loss": 0.3793,
      "step": 960
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0002,
      "loss": 0.503,
      "step": 970
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0002,
      "loss": 0.3353,
      "step": 980
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0002,
      "loss": 0.4212,
      "step": 990
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0002,
      "loss": 0.341,
      "step": 1000
    },
    {
      "epoch": 1.7,
      "eval_loss": 0.4644744396209717,
      "eval_runtime": 102.0988,
      "eval_samples_per_second": 9.794,
      "eval_steps_per_second": 4.897,
      "step": 1000
    },
    {
      "epoch": 1.7,
      "mmlu_eval_accuracy": 0.4532186308687434,
      "mmlu_eval_accuracy_abstract_algebra": 0.18181818181818182,
      "mmlu_eval_accuracy_anatomy": 0.5714285714285714,
      "mmlu_eval_accuracy_astronomy": 0.5,
      "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": 0.25,
      "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": 0.46153846153846156,
      "mmlu_eval_accuracy_econometrics": 0.16666666666666666,
      "mmlu_eval_accuracy_electrical_engineering": 0.375,
      "mmlu_eval_accuracy_elementary_mathematics": 0.4146341463414634,
      "mmlu_eval_accuracy_formal_logic": 0.21428571428571427,
      "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": 0.4444444444444444,
      "mmlu_eval_accuracy_high_school_european_history": 0.5,
      "mmlu_eval_accuracy_high_school_geography": 0.7272727272727273,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.6190476190476191,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.3488372093023256,
      "mmlu_eval_accuracy_high_school_mathematics": 0.10344827586206896,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.4230769230769231,
      "mmlu_eval_accuracy_high_school_physics": 0.29411764705882354,
      "mmlu_eval_accuracy_high_school_psychology": 0.7333333333333333,
      "mmlu_eval_accuracy_high_school_statistics": 0.34782608695652173,
      "mmlu_eval_accuracy_high_school_us_history": 0.5454545454545454,
      "mmlu_eval_accuracy_high_school_world_history": 0.5,
      "mmlu_eval_accuracy_human_aging": 0.7391304347826086,
      "mmlu_eval_accuracy_human_sexuality": 0.3333333333333333,
      "mmlu_eval_accuracy_international_law": 0.7692307692307693,
      "mmlu_eval_accuracy_jurisprudence": 0.2727272727272727,
      "mmlu_eval_accuracy_logical_fallacies": 0.6111111111111112,
      "mmlu_eval_accuracy_machine_learning": 0.36363636363636365,
      "mmlu_eval_accuracy_management": 0.45454545454545453,
      "mmlu_eval_accuracy_marketing": 0.72,
      "mmlu_eval_accuracy_medical_genetics": 0.7272727272727273,
      "mmlu_eval_accuracy_miscellaneous": 0.6744186046511628,
      "mmlu_eval_accuracy_moral_disputes": 0.5263157894736842,
      "mmlu_eval_accuracy_moral_scenarios": 0.24,
      "mmlu_eval_accuracy_nutrition": 0.48484848484848486,
      "mmlu_eval_accuracy_philosophy": 0.47058823529411764,
      "mmlu_eval_accuracy_prehistory": 0.4857142857142857,
      "mmlu_eval_accuracy_professional_accounting": 0.2903225806451613,
      "mmlu_eval_accuracy_professional_law": 0.3352941176470588,
      "mmlu_eval_accuracy_professional_medicine": 0.41935483870967744,
      "mmlu_eval_accuracy_professional_psychology": 0.4492753623188406,
      "mmlu_eval_accuracy_public_relations": 0.5833333333333334,
      "mmlu_eval_accuracy_security_studies": 0.48148148148148145,
      "mmlu_eval_accuracy_sociology": 0.5454545454545454,
      "mmlu_eval_accuracy_us_foreign_policy": 0.5454545454545454,
      "mmlu_eval_accuracy_virology": 0.3888888888888889,
      "mmlu_eval_accuracy_world_religions": 0.7368421052631579,
      "mmlu_loss": 0.9635014412705799,
      "step": 1000
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.0002,
      "loss": 0.4641,
      "step": 1010
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0002,
      "loss": 0.4247,
      "step": 1020
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0002,
      "loss": 0.3628,
      "step": 1030
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0002,
      "loss": 0.3531,
      "step": 1040
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0002,
      "loss": 0.407,
      "step": 1050
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.0002,
      "loss": 0.3457,
      "step": 1060
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.0002,
      "loss": 0.3636,
      "step": 1070
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.0002,
      "loss": 0.4112,
      "step": 1080
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.0002,
      "loss": 0.4043,
      "step": 1090
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.0002,
      "loss": 0.4891,
      "step": 1100
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0002,
      "loss": 0.4216,
      "step": 1110
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.0002,
      "loss": 0.2883,
      "step": 1120
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.0002,
      "loss": 0.4063,
      "step": 1130
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.0002,
      "loss": 0.3683,
      "step": 1140
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.0002,
      "loss": 0.3717,
      "step": 1150
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0002,
      "loss": 0.4374,
      "step": 1160
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.0002,
      "loss": 0.4172,
      "step": 1170
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0002,
      "loss": 0.2856,
      "step": 1180
    },
    {
      "epoch": 2.02,
      "learning_rate": 0.0002,
      "loss": 0.265,
      "step": 1190
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0002,
      "loss": 0.2448,
      "step": 1200
    },
    {
      "epoch": 2.04,
      "eval_loss": 0.5078285336494446,
      "eval_runtime": 102.2139,
      "eval_samples_per_second": 9.783,
      "eval_steps_per_second": 4.892,
      "step": 1200
    },
    {
      "epoch": 2.04,
      "mmlu_eval_accuracy": 0.45134380789496525,
      "mmlu_eval_accuracy_abstract_algebra": 0.2727272727272727,
      "mmlu_eval_accuracy_anatomy": 0.6428571428571429,
      "mmlu_eval_accuracy_astronomy": 0.5,
      "mmlu_eval_accuracy_business_ethics": 0.7272727272727273,
      "mmlu_eval_accuracy_clinical_knowledge": 0.4827586206896552,
      "mmlu_eval_accuracy_college_biology": 0.375,
      "mmlu_eval_accuracy_college_chemistry": 0.25,
      "mmlu_eval_accuracy_college_computer_science": 0.36363636363636365,
      "mmlu_eval_accuracy_college_mathematics": 0.36363636363636365,
      "mmlu_eval_accuracy_college_medicine": 0.45454545454545453,
      "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": 0.25,
      "mmlu_eval_accuracy_electrical_engineering": 0.125,
      "mmlu_eval_accuracy_elementary_mathematics": 0.3902439024390244,
      "mmlu_eval_accuracy_formal_logic": 0.35714285714285715,
      "mmlu_eval_accuracy_global_facts": 0.4,
      "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": 0.7272727272727273,
      "mmlu_eval_accuracy_high_school_government_and_politics": 0.5238095238095238,
      "mmlu_eval_accuracy_high_school_macroeconomics": 0.37209302325581395,
      "mmlu_eval_accuracy_high_school_mathematics": 0.1724137931034483,
      "mmlu_eval_accuracy_high_school_microeconomics": 0.3076923076923077,
      "mmlu_eval_accuracy_high_school_physics": 0.35294117647058826,
      "mmlu_eval_accuracy_high_school_psychology": 0.7333333333333333,
      "mmlu_eval_accuracy_high_school_statistics": 0.2608695652173913,
      "mmlu_eval_accuracy_high_school_us_history": 0.6818181818181818,
      "mmlu_eval_accuracy_high_school_world_history": 0.46153846153846156,
      "mmlu_eval_accuracy_human_aging": 0.6956521739130435,
      "mmlu_eval_accuracy_human_sexuality": 0.25,
      "mmlu_eval_accuracy_international_law": 0.8461538461538461,
      "mmlu_eval_accuracy_jurisprudence": 0.2727272727272727,
      "mmlu_eval_accuracy_logical_fallacies": 0.5,
      "mmlu_eval_accuracy_machine_learning": 0.2727272727272727,
      "mmlu_eval_accuracy_management": 0.45454545454545453,
      "mmlu_eval_accuracy_marketing": 0.68,
      "mmlu_eval_accuracy_medical_genetics": 0.7272727272727273,
      "mmlu_eval_accuracy_miscellaneous": 0.6162790697674418,
      "mmlu_eval_accuracy_moral_disputes": 0.5263157894736842,
      "mmlu_eval_accuracy_moral_scenarios": 0.25,
      "mmlu_eval_accuracy_nutrition": 0.5151515151515151,
      "mmlu_eval_accuracy_philosophy": 0.5,
      "mmlu_eval_accuracy_prehistory": 0.4,
      "mmlu_eval_accuracy_professional_accounting": 0.25806451612903225,
      "mmlu_eval_accuracy_professional_law": 0.34705882352941175,
      "mmlu_eval_accuracy_professional_medicine": 0.3548387096774194,
      "mmlu_eval_accuracy_professional_psychology": 0.463768115942029,
      "mmlu_eval_accuracy_public_relations": 0.5833333333333334,
      "mmlu_eval_accuracy_security_studies": 0.4444444444444444,
      "mmlu_eval_accuracy_sociology": 0.5,
      "mmlu_eval_accuracy_us_foreign_policy": 0.5454545454545454,
      "mmlu_eval_accuracy_virology": 0.3888888888888889,
      "mmlu_eval_accuracy_world_religions": 0.6842105263157895,
      "mmlu_loss": 0.9378224162934421,
      "step": 1200
    }
  ],
  "max_steps": 5000,
  "num_train_epochs": 9,
  "total_flos": 1.1171389466360218e+17,
  "trial_name": null,
  "trial_params": null
}