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

    Neuronpedia

    INDEX
    Explanations

    compound noun phrases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ak
    0.51
    ní
    0.51
    ın
    0.49
     červ
    0.43
    ၚ
    0.42
    et
    0.42
    button
    0.42
    <unused973>
    0.42
    <unused307>
    0.41
    ında
    0.41
    POSITIVE LOGITS
     zelfs
    0.40
    。
    0.36
    🤨
    0.34
     mishaps
    0.31
     luck
    0.30
     incarnations
    0.30
    Said
    0.29
    比如
    0.29
     establishments
    0.29
    OMG
    0.28
    Act Density 0.365%

    No Known Activations