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

    Neuronpedia

    INDEX
    Explanations

    actions and transformations related to physical movement and structure

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    id
    -0.15
    -0.15
    424
    -0.14
    475
    -0.14
     Birch
    -0.14
    442
    -0.14
     interchangeable
    -0.13
    435
    -0.13
    bie
    -0.13
    295
    -0.13
    POSITIVE LOGITS
     обÑĢазованиÑı
    0.19
     sphere
    0.18
     ball
    0.18
     heap
    0.18
     blob
    0.17
     mass
    0.17
    -ball
    0.17
     formations
    0.17
     BALL
    0.17
     haystack
    0.16
    Act Density 0.275%

    No Known Activations