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

    Neuronpedia

    INDEX
    Explanations

    for models

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    č
    0.46
    ⢸
    0.45
    นต์
    0.42
    emies
    0.41
    تباه
    0.40
     Birken
    0.40
    戽
    0.40
    myForm
    0.39
    ুরে
    0.39
     Règles
    0.39
    POSITIVE LOGITS
     restricts
    0.47
     deactivate
    0.46
     quantify
    0.44
     inactive
    0.43
     seasoned
    0.43
     thanks
    0.42
     scratched
    0.41
     stiff
    0.41
     reaffirmed
    0.41
     coalesce
    0.41
    Act Density 0.000%

    No Known Activations