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

    Neuronpedia

    INDEX
    Explanations

    content splitting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    呗
    0.65
     alá
    0.63
    <unused88>
    0.63
     نفسك
    0.60
    PUS
    0.58
    ベート
    0.58
     adapter
    0.57
     περι
    0.57
     Framework
    0.57
     suppos
    0.57
    POSITIVE LOGITS
    offrey
    0.56
     splitting
    0.55
     emeritus
    0.54
     বিভক্ত
    0.53
    离开
    0.52
     splits
    0.52
    ai
    0.52
    splits
    0.50
    omeres
    0.50
    Mature
    0.50
    Act Density 0.257%

    No Known Activations