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

    Neuronpedia

    INDEX
    Explanations

    component descriptions that focus on physical attributes or actions

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

    punctuation marks, especially periods and commas, that conclude narrative sentences or paragraphs.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    conversational narrative passages with dialogue and descriptive details about characters and their interactions.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    New Auto-Interp
    Negative Logits
     esso
    -0.66
     dets
    -0.63
    ньому
    -0.54
     []*
    -0.48
    byshev
    -0.44
     Económica
    -0.43
    ILLES
    -0.42
    valo
    -0.41
    DOCTYPE
    -0.41
    ladin
    -0.41
    POSITIVE LOGITS
     she
    3.39
     her
    2.77
     그녀
    2.67
     hennes
    2.39
    她
    2.36
    彼女は
    2.31
    彼女の
    2.31
     její
    2.31
    she
    2.19
    她的
    2.19
    Act Density 1.318%

    No Known Activations