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

    Neuronpedia

    INDEX
    Explanations

    parent

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    (defun
    -0.07
     Md
    -0.07
    rooms
    -0.07
    _instruction
    -0.06
    676
    -0.06
    _SW
    -0.06
    WORDS
    -0.06
    _negative
    -0.06
    평
    -0.06
    _predictions
    -0.06
    POSITIVE LOGITS
    ction
    0.08
    lovak
    0.06
     Strength
    0.06
     крови
    0.06
     experiment
    0.06
     сті
    0.06
    iph
    0.06
    .Rollback
    0.06
     Messenger
    0.06
    .tagName
    0.05
    Act Density 0.009%

    No Known Activations