© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 91932
    Prev
    Next
    INDEX
    Explanations

    conjunctions and qualifiers that introduce contrasting or mitigating statements

    oai_token-act-pair · gpt-4o-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
    .hw
    -0.08
    mour
    -0.08
    ocre
    -0.08
    istrovstvÃŃ
    -0.08
    عب
    -0.08
     пÑĢавило
    -0.08
    Äįer
    -0.08
    olie
    -0.08
    uil
    -0.08
    ÐŀÐĴ
    -0.08
    POSITIVE LOGITS
    /or
    0.10
    ;
    0.07
    /of
    0.07
    verts
    0.07
    âĤ¬“
    0.06
    ,
    0.06
    Ìģ
    0.06
    /OR
    0.06
    ...
    0.06
    ients
    0.06
    Activations Density 0.165%

    No Known Activations