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

    Neuronpedia

    INDEX
    Explanations

    the word "in" across various contexts, suggesting a focus on locational or situational references

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Huguen
    -0.67
    紹介します
    -0.67
    poptosis
    -0.67
    狙
    -0.64
     Euphrates
    -0.63
     eiffel
    -0.63
     Silurian
    -0.62
     Chriftian
    -0.62
     Hadrian
    -0.61
    Viitteet
    -0.60
    POSITIVE LOGITS
     terms
    1.25
     the
    1.22
     accordance
    1.17
     relation
    1.12
     spite
    1.07
     lieu
    1.04
     regards
    1.03
     conjunction
    1.02
     order
    0.97
     addition
    0.96
    Act Density 1.156%

    No Known Activations