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

    Neuronpedia

    INDEX
    Explanations

    English to transformation, engineering

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     overwhelming
    2.13
    隹
    1.86
     hydrocarbons
    1.86
     inclement
    1.84
     inhibit
    1.82
    ordinating
    1.76
     intent
    1.75
     overwhelm
    1.74
     alimentação
    1.73
     ï
    1.71
    POSITIVE LOGITS
    на
    2.94
    م
    2.83
    м
    2.71
    ड
    2.29
    ский
    2.21
    woman
    2.07
    ು
    2.06
    म
    2.04
    y
    1.98
    ν
    1.97
    Act Density 0.246%

    No Known Activations