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

    Neuronpedia

    INDEX
    Explanations

    configuration and code keywords

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    IRCONIUM
    0.35
    DeathRecord
    0.32
     наші
    0.30
    كلمنا
    0.30
    StillWater
    0.30
    ัพท์
    0.30
     causando
    0.29
    اع
    0.29
    yscrapers
    0.29
    Geschichte
    0.29
    POSITIVE LOGITS
    _
    0.41
     and
    0.38
     .
    0.37
    ="
    0.37
     to
    0.36
     t
    0.36
     =
    0.35
     from
    0.35
            
    0.35
     include
    0.34
    Act Density 0.050%

    No Known Activations