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

    Neuronpedia

    INDEX
    Explanations

    references to suspicious activities or threats

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ijken
    -0.17
    ietet
    -0.15
     Collision
    -0.14
    ouch
    -0.13
    lag
    -0.13
    collision
    -0.13
    rezent
    -0.13
    .friend
    -0.13
    дап
    -0.13
    æŀ
    -0.12
    POSITIVE LOGITS
     threats
    0.30
     threat
    0.28
     hoax
    0.28
     threatening
    0.27
     bomb
    0.25
     evacuated
    0.24
     Threat
    0.24
    -threat
    0.24
     evac
    0.23
     suspicious
    0.22
    Act Density 0.031%

    No Known Activations