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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. GPT2-Small
    3. Transcoders Residuals
    4. 8-TRES-DC
    5. 190
    Prev
    Next
    INDEX
    Explanations

    punctuation marks, particularly periods and commas, as well as their context in sentences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    issance
    -0.68
    neau
    -0.65
     curtains
    -0.64
     sle
    -0.64
    urat
    -0.63
     attain
    -0.62
     owes
    -0.59
     eyebrows
    -0.59
     lounge
    -0.58
     paran
    -0.58
    POSITIVE LOGITS
     DOI
    0.88
    RANT
    0.71
    orio
    0.69
    enting
    0.67
    ={
    0.65
    ented
    0.63
     fixme
    0.62
     ACPI
    0.62
    Acknowled
    0.62
    cot
    0.62
    Activations Density 0.208%

    No Known Activations