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

    Neuronpedia

    INDEX
    Explanations

    concepts and states

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     qualifies
    0.36
    <unused2173>
    0.36
     বেশিরভাগ
    0.36
     इंपोर्टेंट
    0.35
     avoided
    0.35
     selects
    0.34
     alternates
    0.34
    োই
    0.34
     variés
    0.34
     athletes
    0.33
    POSITIVE LOGITS
     которую
    0.45
     한
    0.44
     Without
    0.43
     그것
    0.42
     microcosm
    0.42
     profundo
    0.41
     koju
    0.40
    忿
    0.40
     ו
    0.40
     וא
    0.40
    Act Density 0.216%

    No Known Activations