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

    Neuronpedia

    INDEX
    Explanations

    words signaling a contrast or contradiction in a sentence

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Yet."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    76561
    -0.75
    heads
    -0.74
    spir
    -0.73
    strings
    -0.71
    packs
    -0.68
    tains
    -0.65
    units
    -0.65
    mens
    -0.64
    ricular
    -0.63
    tein
    -0.63
    POSITIVE LOGITS
    tons
    0.86
    theless
    0.82
    heric
    0.78
     alas
    0.78
     nevertheless
    0.73
     nonetheless
    0.72
     Cors
    0.70
     somehow
    0.68
     Nguyen
    0.66
     despite
    0.66
    Act Density 0.011%

    No Known Activations