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

    Neuronpedia

    INDEX
    Explanations

    references to technical setup options and configurations

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

    Syntax or code structure that shows relationships between elements, often in programming contexts. This includes variable declarations, method references, and structural relationships like containment or modification indicated by prepositions ("in", "from", "by", "to") and punctuation.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    New Auto-Interp
    Negative Logits
    --------------------------------------------------------------------------↵
    -0.07
     anh
    -0.07
     subj
    -0.07
    (.)
    -0.07
    .accounts
    -0.07
    ãĥ©ãĥ³ãĥī
    -0.07
    Ñīина
    -0.07
    .)↵↵↵↵
    -0.07
    .getElements
    -0.07
    __/
    -0.07
    POSITIVE LOGITS
    inqu
    0.07
    lier
    0.06
    PasswordEncoder
    0.06
     Schro
    0.06
    gua
    0.06
    Ñĵ
    0.06
     Qgs
    0.06
    ewise
    0.06
    bout
    0.06
    addtogroup
    0.06
    Act Density 1.016%

    No Known Activations