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

    Neuronpedia

    INDEX
    Explanations

    specific technical or code-related terms and variables

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    empt
    -0.17
    aque
    -0.16
     Rae
    -0.15
    udit
    -0.14
    aal
    -0.14
    umen
    -0.14
     jewel
    -0.14
    une
    -0.13
    inder
    -0.13
    acos
    -0.13
    POSITIVE LOGITS
    ations
    0.26
    ated
    0.25
    ating
    0.25
    ing
    0.24
    ation
    0.20
    ative
    0.20
    able
    0.19
    ingen
    0.19
    ate
    0.19
    acja
    0.19
    Act Density 0.204%

    No Known Activations