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

    Neuronpedia

    INDEX
    Explanations

    recursive queries and old nordish

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    9
    0.55
    gar
    0.52
    f
    0.51
    4
    0.51
    ed
    0.50
    (
    0.50
    le
    0.50
    ing
    0.48
    yka
    0.46
    al
    0.46
    POSITIVE LOGITS
    ților
    0.53
     менен
    0.52
     жүкт
    0.50
     गोद
    0.50
    팸
    0.50
    प्या
    0.49
     जड़ों
    0.48
     thủy
    0.47
     ойноо
    0.47
     mái
    0.46
    Act Density 0.000%

    No Known Activations