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

    Neuronpedia

    INDEX
    Explanations

    contrasting conjunctions and phrases that indicate complexity or nuance in relationships between ideas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     not
    -0.35
     NOT
    -0.33
     không
    -0.29
     не
    -0.27
     nicht
    -0.27
    ä¸į
    -0.26
     niet
    -0.26
     Not
    -0.24
    not
    -0.24
     tidak
    -0.24
    POSITIVE LOGITS
     rather
    0.39
    rather
    0.32
     Rather
    0.32
    Rather
    0.30
     بÙĦÚ©Ùĩ
    0.25
     actually
    0.25
     sondern
    0.22
     importantly
    0.20
     plutôt
    0.19
     quite
    0.18
    Act Density 0.098%

    No Known Activations