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

    Neuronpedia

    INDEX
    Explanations

    adjectives related to physical attributes or actions

    oai_token-act-pair · gpt-3.5-turbo

    positive adjectives and adverbs that convey a sense of safety or improvement

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ITNESS
    -0.74
    vous
    -0.72
    Quantity
    -0.68
    iph
    -0.67
    âĢ¢âĢ¢âĢ¢âĢ¢
    -0.65
     Divinity
    -0.64
    akings
    -0.59
    ROR
    -0.58
    ãĥĥãĥĪ
    -0.55
     Difference
    -0.55
    POSITIVE LOGITS
    aneously
    1.29
    heartedly
    1.09
    handedly
    1.05
     enough
    1.03
    ly
    1.01
    edly
    0.94
    istically
    0.90
    rily
    0.90
    lly
    0.87
     distances
    0.84
    Act Density 0.250%

    No Known Activations