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

    Neuronpedia

    INDEX
    Explanations

    adjectives related to excessive or extreme behavior

    oai_token-act-pair · gpt-3.5-turbo

    negative modifiers that suggest an excessive or overwhelming quality

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    blood
    -0.70
    tor
    -0.68
     Dispatch
    -0.65
    Life
    -0.65
    bard
    -0.64
     Scene
    -0.64
    Unity
    -0.62
    brate
    -0.61
    iamond
    -0.60
    cloth
    -0.60
    POSITIVE LOGITS
     overly
    3.11
     excessively
    2.19
     unnecessarily
    1.77
     excessive
    1.62
     overc
    1.53
     unrealistic
    1.50
     undue
    1.48
     overe
    1.46
     inappropriately
    1.44
     unreasonable
    1.41
    Act Density 0.033%

    No Known Activations