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

    Neuronpedia

    INDEX
    Explanations

    idioms and figurative language

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Um
    0.66
     Teknologi
    0.59
    UM
    0.58
     robotic
    0.56
    U
    0.56
     beep
    0.55
    Uh
    0.55
     are
    0.54
     endeavor
    0.52
     Media
    0.51
    POSITIVE LOGITS
     idiom
    0.59
    ст
    0.56
    称
    0.54
    п
    0.53
    рт
    0.52
     idioms
    0.52
     phrases
    0.51
    称为
    0.49
    ристо
    0.48
    composition
    0.48
    Act Density 0.219%

    No Known Activations