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

    Neuronpedia

    INDEX
    Explanations

    conditional statements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    go
    0.49
    í
    0.49
    by
    0.48
    for
    0.48
    rän
    0.48
    shooting
    0.47
    {
    0.47
    firing
    0.46
    lom
    0.46
    if
    0.46
    POSITIVE LOGITS
     Wooden
    0.48
    চৈতন্যের
    0.48
     اسی
    0.47
    アルミ
    0.46
     Arnold
    0.46
     натура
    0.46
     deductible
    0.44
     wealthier
    0.44
     Raymond
    0.44
     힘들
    0.44
    Act Density 0.002%

    No Known Activations