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

    Neuronpedia

    INDEX
    Explanations

    before, after, technical terms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Class
    0.53
     limits
    0.51
     mengatasi
    0.47
    nabi
    0.47
     claws
    0.47
     conscient
    0.46
     scl
    0.45
    ombang
    0.45
    agory
    0.45
     limitations
    0.45
    POSITIVE LOGITS
    THE
    0.43
     перед
    0.43
    Перед
    0.42
    选
    0.42
    keyDown
    0.42
    事前
    0.41
    the
    0.40
    降り
    0.40
    ട്ടില്‍
    0.40
    燄
    0.40
    Act Density 0.008%

    No Known Activations