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

    Neuronpedia

    INDEX
    Explanations

    what is/are X

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    startsWith
    0.84
    श्वर
    0.80
    Deutschland
    0.78
    ಗರ
    0.78
    هم
    0.77
     није
    0.76
    Perf
    0.75
     защиту
    0.75
    ك
    0.73
    perf
    0.73
    POSITIVE LOGITS
     supposed
    1.31
     liable
    1.15
     positioned
    1.07
     located
    1.06
     situated
    1.00
     meant
    0.99
     perceived
    0.98
     capable
    0.98
     during
    0.97
     stationed
    0.96
    Act Density 0.090%

    No Known Activations