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

    Neuronpedia

    INDEX
    Explanations

    references to demonstrative pronouns and related expressions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    "that" preceding descriptive words

    np_acts-logits-general · gemini-2.0-flash

    that followed by descriptive word

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     PLWABN
    -0.65
     Whigs
    -0.61
     motifs
    -0.54
     regiments
    -0.54
    出版年
    -0.51
     Sikhs
    -0.51
     lesbians
    -0.51
     parishes
    -0.51
     geslacht
    -0.50
     installments
    -0.50
    POSITIVE LOGITS
     wonderful
    0.87
     extra
    0.85
     pesky
    0.84
     big
    0.82
     little
    0.79
     amazing
    0.79
     huge
    0.78
     precious
    0.75
     vast
    0.75
     great
    0.74
    Act Density 0.164%

    No Known Activations