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

    Neuronpedia

    INDEX
    Explanations

    conjunctions and commas

    oai_token-act-pair · gpt-3.5-turbo

    punctuation and conjunctions that indicate a contrast or connection in ideas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    tained
    -0.55
    venth
    -0.55
    aughtered
    -0.53
     guiActiveUn
    -0.52
    acket
    -0.52
    ulla
    -0.50
     ÂŃ
    -0.48
     Tours
    -0.48
     Partnership
    -0.48
    elfth
    -0.47
    POSITIVE LOGITS
     huh
    1.08
     etc
    0.93
     please
    0.91
     eh
    0.91
     yeah
    0.88
     dont
    0.87
     doesnt
    0.85
     preferably
    0.80
     but
    0.76
     tho
    0.75
    Act Density 0.415%

    No Known Activations