Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    official declarations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     incompre
    0.64
    Trusted
    0.60
     troubled
    0.59
     kidney
    0.59
     generosity
    0.59
     dispute
    0.58
     Trusted
    0.58
    υχ
    0.58
     fraud
    0.57
     trustworthiness
    0.57
    POSITIVE LOGITS
     officially
    4.11
     formally
    4.06
     formal
    3.74
     официально
    3.54
     official
    3.46
     Officially
    3.40
    正式
    3.36
     औपचारिक
    3.28
    Formal
    3.28
     oficialmente
    3.27
    Act Density 0.408%

    No Known Activations