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

    Neuronpedia

    INDEX
    Explanations

    technical or conditional circumstances

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     g
    0.44
    V
    0.44
     Gold
    0.41
     accompan
    0.41
     gold
    0.40
     V
    0.40
     prove
    0.39
    Gold
    0.39
     source
    0.38
     generate
    0.38
    POSITIVE LOGITS
    KIS
    0.46
     нрав
    0.45
    nungs
    0.44
    कोर
    0.43
    が好き
    0.41
     विद्यार्थी
    0.40
    ậc
    0.40
    വീ
    0.39
    aksana
    0.39
    ドリンク
    0.39
    Act Density 0.000%

    No Known Activations