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

    Neuronpedia

    INDEX
    Explanations

    multimodal LLMs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     speci
    0.59
     kaks
    0.49
    ස
    0.48
     exclu
    0.48
    鵑
    0.48
     tying
    0.48
     tangents
    0.48
    jeta
    0.45
     insectes
    0.45
     consideramos
    0.45
    POSITIVE LOGITS
    ורי
    0.50
    记得
    0.49
    寓
    0.47
    IMM
    0.44
    众多
    0.44
    汝
    0.43
    时光
    0.42
    GOING
    0.42
    Ng
    0.41
    来
    0.41
    Act Density 0.001%

    No Known Activations