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

    Neuronpedia

    INDEX
    Explanations

    junction table

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Attachment
    0.69
     Attach
    0.63
     attachment
    0.60
    附件
    0.60
     attachments
    0.59
    attachment
    0.55
    Attachment
    0.55
    Attachments
    0.54
     Attached
    0.52
     attach
    0.50
    POSITIVE LOGITS
    Eras
    0.39
    cleaned
    0.37
    збе
    0.37
    claw
    0.37
    Am
    0.37
    POR
    0.37
     claw
    0.37
    Mw
    0.36
    Cons
    0.36
    Bismillahirrah
    0.36
    Act Density 0.000%

    No Known Activations