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

    Neuronpedia

    INDEX
    Explanations

    sentences with first-person pronouns followed by statements

    oai_token-act-pair · gpt-3.5-turbo

    multiple occurrences of periods, indicating the end of sentences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ishes
    -0.70
     spawning
    -0.62
    Þ
    -0.60
     dash
    -0.59
     confidentiality
    -0.59
     bleach
    -0.58
    HAHA
    -0.57
     aborted
    -0.56
     webcam
    -0.56
     fragrance
    -0.56
    POSITIVE LOGITS
    imgur
    0.82
    Lenin
    0.80
     Lenin
    0.77
    seed
    0.72
    sbm
    0.72
    medium
    0.69
    material
    0.67
    cloud
    0.67
    spir
    0.67
    MX
    0.65
    Act Density 0.051%

    No Known Activations