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

    Neuronpedia

    INDEX
    Explanations

    phrases related to negative impacts or blows, particularly in a figurative sense

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    é¬
    -0.18
     Bulk
    -0.15
    ctors
    -0.15
    imple
    -0.14
    Äįást
    -0.14
    laÄį
    -0.14
    ÏĢÎŃ
    -0.14
    Bulk
    -0.14
    esub
    -0.14
    itchen
    -0.14
    POSITIVE LOGITS
     blow
    0.74
     blows
    0.66
     Blow
    0.62
     punches
    0.39
     punch
    0.38
     hits
    0.36
     blew
    0.35
     hit
    0.34
     ÑĥдаÑĢ
    0.32
     blown
    0.32
    Act Density 0.172%

    No Known Activations