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

    Neuronpedia

    INDEX
    Explanations

    parchment

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Adaptive
    0.40
     замі
    0.40
     relay
    0.39
    🚔
    0.39
     Lieutenant
    0.38
     الثاني
    0.38
    Physics
    0.38
     Lieut
    0.37
    Heritage
    0.36
     crawling
    0.36
    POSITIVE LOGITS
     parchment
    0.44
     mane
    0.41
    egger
    0.41
     recognise
    0.38
     impec
    0.38
    ﺽ
    0.37
    ridor
    0.37
     erro
    0.37
     Parch
    0.37
    盏
    0.36
    Act Density 0.001%

    No Known Activations