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

    Neuronpedia

    INDEX
    Explanations

    references to explosive events or conditions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    vecs
    -0.16
     Cyr
    -0.15
    itsu
    -0.15
    idth
    -0.15
    enville
    -0.14
    å¹³æĪIJ
    -0.14
    ells
    -0.14
     Cast
    -0.14
    997
    -0.14
    Calibri
    -0.14
    POSITIVE LOGITS
     explosive
    0.30
     Explos
    0.28
     explosives
    0.28
     explosion
    0.26
     Explosion
    0.24
    çĪĨ
    0.24
     ignition
    0.23
    exp
    0.22
     TNT
    0.22
     explode
    0.22
    Act Density 0.066%

    No Known Activations