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

    Neuronpedia

    INDEX
    Explanations

    ffmpeg commands

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Breath
    0.46
    voice
    0.46
    Breath
    0.45
     Respiratory
    0.43
    Speech
    0.43
    语音
    0.42
    圄
    0.42
     giọng
    0.42
    录
    0.42
    聲
    0.41
    POSITIVE LOGITS
     conic
    0.46
     server
    0.44
     poll
    0.42
     controls
    0.42
     alkali
    0.42
     rewind
    0.40
     ahead
    0.40
     thorium
    0.40
     దా
    0.40
     folly
    0.40
    Act Density 0.129%

    No Known Activations