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

    Neuronpedia

    INDEX
    Explanations

    identifying "you are" / "you speak" / "give me"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    anson
    -0.09
    agination
    -0.08
     whit
    -0.08
     Blowjob
    -0.08
    intros
    -0.08
    à¸ģารà¸ĵ
    -0.08
    Rua
    -0.08
    SPATH
    -0.08
    ARING
    -0.07
    embros
    -0.07
    POSITIVE LOGITS
    #ab
    0.10
    ify
    0.09
    .gov
    0.09
    ese
    0.09
    ia
    0.09
     OnTrigger
    0.09
    fully
    0.08
    каж
    0.08
    ually
    0.08
    enberg
    0.08
    Act Density 0.187%

    No Known Activations