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

    Neuronpedia

    INDEX
    Explanations

    phrases related to explosive or impactful actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ummings
    -0.16
    olland
    -0.16
    ully
    -0.15
    untime
    -0.14
    ABCDEFGHIJKLMNOP
    -0.13
    merce
    -0.13
    amerate
    -0.13
    lest
    -0.13
    anel
    -0.13
    óa
    -0.13
    POSITIVE LOGITS
    torch
    0.36
     smoke
    0.27
     apart
    0.26
     kisses
    0.25
     blew
    0.24
     bubbles
    0.24
     torch
    0.23
     away
    0.23
    out
    0.23
    tor
    0.22
    Act Density 0.017%

    No Known Activations