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

    Neuronpedia

    INDEX
    Explanations

    instances of punctuation marks

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

    negative exclamations and intensifiers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     providing
    -0.63
     provides
    -0.59
     provide
    -0.55
     highly
    -0.53
     utilizing
    -0.51
     אשר
    -0.50
     presented
    -0.49
     enabling
    -0.48
     primarily
    -0.48
     subsequent
    -0.48
    POSITIVE LOGITS
     mierda
    0.65
     shitty
    0.65
     jeito
    0.64
     sowas
    0.63
     fuckin
    0.62
     crappy
    0.61
     houſe
    0.59
     goddamn
    0.59
     fuck
    0.57
    fuck
    0.57
    Act Density 0.166%

    No Known Activations