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

    Neuronpedia

    INDEX
    Explanations

    words related to minerals and elements

    oai_token-act-pair · gpt-3.5-turbo

    references to quantities and classifications of elements or minerals

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Hollande
    -0.66
    Fear
    -0.66
    deen
    -0.64
    Soon
    -0.63
    Pitt
    -0.62
    oward
    -0.61
     woke
    -0.60
    Dam
    -0.59
    Suddenly
    -0.59
    Putin
    -0.59
    POSITIVE LOGITS
     ranges
    0.92
     descriptions
    0.90
     varies
    0.88
     specialize
    0.86
     grouped
    0.85
     categorized
    0.84
     definitions
    0.83
     specialization
    0.82
     combinations
    0.81
     vary
    0.81
    Act Density 1.099%

    No Known Activations