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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating contrast or contradiction

    oai_token-act-pair · gpt-3.5-turbo

    conjunctions and transitional phrases that indicate contrast or connection

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ilian
    -0.65
    76561
    -0.64
    ulkan
    -0.63
    pmwiki
    -0.62
    Dispatch
    -0.62
    lass
    -0.61
    ject
    -0.60
    ãĤ©
    -0.58
    ioxide
    -0.58
    ãĤ¼ãĤ¦ãĤ¹
    -0.56
    POSITIVE LOGITS
     for
    1.45
    for
    1.19
     For
    1.05
     FOR
    1.01
    For
    0.96
    FOR
    0.91
     Against
    0.81
     against
    0.77
    against
    0.71
     especially
    0.66
    Act Density 0.245%

    No Known Activations