© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    5. Llama3.1-8B (Base)
    7. Residual Stream
    9. 17-LLAMASCOPE-RES-131K
    11. 42813
    Prev
    Next
    INDEX
    Explanations

    punctuation marks, specifically quotation marks and apostrophes used in dialogues or titles

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
    ÌĨ
    -0.10
     ÑģобоÑİ
    -0.09
    #ab
    -0.08
    ÑĢип
    -0.08
    #ac
    -0.08
    .reporting
    -0.08
    rana
    -0.08
    ---</
    -0.08
    лаз
    -0.08
    #ad
    -0.08
    POSITIVE LOGITS
    ish
    0.07
    ong
    0.07
    as
    0.06
    ima
    0.06
    ism
    0.06
    ans
    0.06
    al
    0.06
    Question
    0.06
    j
    0.06
    ins
    0.06
    Activations Density 0.014%

    No Known Activations