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

    Neuronpedia

    INDEX
    Explanations

    words related to expressing strong negative opinions or emotions, especially disgust or repulsion

    oai_token-act-pair · gpt-3.5-turbo

    negative adjectives describing unpleasantness or moral repugnance

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    inoa
    -0.77
    arist
    -0.76
    aran
    -0.75
    arta
    -0.74
    ingham
    -0.74
    essor
    -0.73
    arat
    -0.71
    ilogy
    -0.71
    plane
    -0.68
    olin
    -0.68
    POSITIVE LOGITS
     disgusting
    1.02
     despicable
    0.91
     vile
    0.78
    ritch
    0.74
    Magikarp
    0.73
     disgrace
    0.72
     creatures
    0.70
     reptiles
    0.70
     racists
    0.68
     pedoph
    0.68
    Act Density 0.018%

    No Known Activations