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

    Neuronpedia

    INDEX
    Explanations

    the letter "d" preceded or followed by specific letters

    oai_token-act-pair · gpt-3.5-turbo

    signals or structural markers indicating the end of a document or a significant transition

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    EStream
    -0.80
    ©¶æ¥µ
    -0.80
    enhagen
    -0.79
    ħĭ
    -0.79
     Inquisitor
    -0.74
     Puzzles
    -0.73
     Chaser
    -0.71
    姫
    -0.71
     ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³
    -0.70
    İĭ
    -0.70
    POSITIVE LOGITS
    arc
    1.03
    ounded
    0.96
    itches
    0.94
    agn
    0.93
    aint
    0.93
    unk
    0.92
    arr
    0.91
    psc
    0.90
    umped
    0.90
    oked
    0.90
    Act Density 0.133%

    No Known Activations