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

    Neuronpedia

    INDEX
    Explanations

    knows smoking

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    match
    0.59
    bow
    0.59
    projection
    0.59
    embeddings
    0.58
    dough
    0.58
    blend
    0.57
    wap
    0.56
    🥨
    0.56
    memcpy
    0.56
    shed
    0.56
    POSITIVE LOGITS
    攽
    0.73
    ел
    0.67
     distra
    0.66
    మయ్య
    0.64
     Mfg
    0.62
    欹
    0.62
     ஹீ
    0.61
     veditabbo
    0.60
     soci
    0.59
     свого
    0.59
    Act Density 0.001%

    No Known Activations