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

    Neuronpedia

    INDEX
    Explanations

    phrases related to revealing or uncovering hidden information

    oai_token-act-pair · gpt-3.5-turbo

    references to curtains or veils, often indicating concealment or secrecy

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ctic
    -0.77
    ishops
    -0.66
    versions
    -0.65
    ciating
    -0.64
    OTAL
    -0.63
     Bald
    -0.61
    etically
    -0.61
    mits
    -0.60
    protein
    -0.59
    ogh
    -0.59
    POSITIVE LOGITS
     curtain
    1.49
     curtains
    1.25
    glass
    0.96
     veil
    0.89
     Veil
    0.82
    AGE
    0.76
    ãĤ¦ãĤ¹
    0.74
    cloth
    0.74
    bridge
    0.74
     rod
    0.70
    Act Density 0.007%

    No Known Activations