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

    Neuronpedia

    INDEX
    Explanations

    words related to exaggerated attention or promotion

    oai_token-act-pair · gpt-3.5-turbo

    mentions of "hype" and legal terms such as "injunction."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     diplomacy
    -0.73
    itime
    -0.71
     Apprentice
    -0.65
    inking
    -0.64
    lected
    -0.63
    aunch
    -0.63
     floats
    -0.62
    fully
    -0.62
    ampunk
    -0.61
     spheres
    -0.61
    POSITIVE LOGITS
     fuss
    1.48
    jee
    1.42
     hype
    1.42
    closure
    1.28
    izu
    1.26
     stigma
    1.22
     injunction
    1.21
    tree
    1.17
     KR
    1.16
    icted
    1.10
    Act Density 0.040%

    No Known Activations