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

    Neuronpedia

    INDEX
    Explanations

    intensifiers or descriptors that convey strong emotions or states of being

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    utterly negative descriptors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     zijne
    -0.67
     auroit
    -0.65
     avoient
    -0.63
     propOrder
    -0.62
     étoient
    -0.60
     étoit
    -0.60
     enfans
    -0.59
     dieß
    -0.59
     queſta
    -0.57
     koszulka
    -0.57
    POSITIVE LOGITS
     quintessential
    0.81
     drenched
    0.60
     exquisitely
    0.59
     immensely
    0.53
     effortlessly
    0.52
     tremendously
    0.52
     exquisite
    0.51
     quint
    0.51
     Quint
    0.49
     intimately
    0.49
    Act Density 0.010%

    No Known Activations