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

    Neuronpedia

    INDEX
    Explanations

    elements of dialogue or quotations in text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    "Yeah
    -0.21
    Ok
    -0.17
     Dude
    -0.17
     ok
    -0.16
    Spoiler
    -0.15
     Ok
    -0.15
    resenter
    -0.15
    okay
    -0.15
    imos
    -0.14
    .ok
    -0.14
    POSITIVE LOGITS
     oh
    0.26
     Oh
    0.25
     Ah
    0.23
     ah
    0.23
     pray
    0.22
    Oh
    0.20
     indeed
    0.20
     truly
    0.19
    oh
    0.18
     heavens
    0.18
    Act Density 0.275%

    No Known Activations