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

    Neuronpedia

    INDEX
    Explanations

    conjunctions followed by a contrasting statement

    oai_token-act-pair · gpt-3.5-turbo

    the conjunction "But" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    heads
    -0.76
     segment
    -0.72
     pier
    -0.63
    cloth
    -0.63
    obyl
    -0.59
    ¯¯¯¯
    -0.58
    sense
    -0.58
     ceremony
    -0.58
     srfN
    -0.57
    ãģ®
    -0.57
    POSITIVE LOGITS
    tons
    1.24
     alas
    0.95
    romeda
    0.83
    theless
    0.82
    chers
    0.80
     luckily
    0.77
    ts
    0.77
    LER
    0.74
    chery
    0.73
     hey
    0.72
    Act Density 0.100%

    No Known Activations