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

    Neuronpedia

    INDEX
    Explanations

    sections of code that have multiple underscores or whitespace characters

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     chicos
    -0.53
     &___
    -0.48
    vol
    -0.47
     исте
    -0.46
    board
    -0.46
    Toole
    -0.46
     văn
    -0.45
     partners
    -0.45
     мол
    -0.45
    بال
    -0.45
    POSITIVE LOGITS
     Савезне
    0.86
    ConstraintMaker
    0.85
    protoimpl
    0.85
     '\\;'
    0.80
    AndEndTag
    0.76
     Roskov
    0.76
    تقاوى
    0.74
    OGND
    0.74
     BorderSide
    0.74
     ErrIntOverflow
    0.72
    Act Density 0.557%

    No Known Activations