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

    Neuronpedia

    INDEX
    Explanations

    instances of engraved or carved elements in descriptions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ãĥ³ãĤ¸
    -0.17
     Stap
    -0.16
    oj
    -0.15
    street
    -0.15
     Lips
    -0.15
    656
    -0.14
     Chern
    -0.13
    izon
    -0.13
     Heard
    -0.13
     street
    -0.13
    POSITIVE LOGITS
     carved
    0.40
     carving
    0.39
     carve
    0.38
    åĪ»
    0.27
     engr
    0.27
     et
    0.27
     khắc
    0.27
     relief
    0.26
     Relief
    0.25
     inscription
    0.25
    Act Density 0.059%

    No Known Activations