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

    Neuronpedia

    INDEX
    Explanations

    significant punctuation and conjunctions that indicate shifts in argument or emphasis

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Instead, Rather, but

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    RenderAtEndOf
    -0.39
    abstractmethod
    -0.32
    endwhile
    -0.32
     CGRectMake
    -0.31
     argint
    -0.30
    ArgumentParser
    -0.29
     Ähnliche
    -0.28
     endwhile
    -0.27
     sagesse
    -0.27
    中です
    -0.27
    POSITIVE LOGITS
     يتيمه
    0.63
    ChildScrollView
    0.57
     sondern
    0.57
     بلکه
    0.56
     ErrIntOverflow
    0.55
    ExtendWith
    0.54
    RunWith
    0.52
     simply
    0.51
     referenties
    0.50
     justru
    0.50
    Act Density 0.264%

    No Known Activations