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

    Neuronpedia

    INDEX
    Explanations

    suggests reliance or avoid talk

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    អារ
    0.46
     関
    0.44
    緩和
    0.43
    ressant
    0.40
     সম্প
    0.38
     immunosupp
    0.38
     સંબંધ
    0.38
    atenin
    0.38
    堑
    0.38
    性和
    0.37
    POSITIVE LOGITS
     ৭৮
    0.39
    Під
    0.38
     vé
    0.38
    लेख
    0.38
     Nobel
    0.38
     Hé
    0.36
    èg
    0.35
     thắng
    0.35
     takiego
    0.34
     👍
    0.34
    Act Density 0.001%

    No Known Activations