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

    Neuronpedia

    INDEX
    Explanations

    asking you to tell

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     unspoken
    0.74
     additions
    0.70
     add
    0.69
    かも
    0.69
    添加到
    0.69
    わかる
    0.68
    Add
    0.67
    额外
    0.67
     ලැබේ
    0.66
    add
    0.66
    POSITIVE LOGITS
     said
    1.26
     told
    1.11
    said
    1.06
     сказал
    0.97
     criticised
    0.89
     satd
    0.89
     sagte
    0.88
     сказала
    0.84
     dijo
    0.84
     questioned
    0.83
    Act Density 0.014%

    No Known Activations