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

    Neuronpedia

    INDEX
    Explanations

    backticks template literals (` `)

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     THREE
    0.57
    ΗΣ
    0.55
     occurring
    0.53
     McConnell
    0.53
    zione
    0.51
     हों
    0.50
     библиоте
    0.50
    ющихся
    0.49
     سوم
    0.49
    鶚
    0.48
    POSITIVE LOGITS
    🔒
    0.52
    াইল
    0.46
     Spacer
    0.44
     दह
    0.44
     idk
    0.43
    景色
    0.43
    ‼️
    0.42
    🥹
    0.41
    🎧
    0.41
     gấp
    0.40
    Act Density 0.001%

    No Known Activations