© 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. 27422
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "useful" or its variations, indicating a focus on practicality or helpfulness

    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
    edb
    -0.08
     usual
    -0.07
    raz
    -0.07
    ilet
    -0.07
    /post
    -0.07
    .nz
    -0.07
    üzel
    -0.07
    rosso
    -0.07
    izen
    -0.07
    olley
    -0.07
    POSITIVE LOGITS
    fully
    0.10
    /help
    0.09
    FUL
    0.08
    lest
    0.08
    lessly
    0.08
     ÃŃch
    0.08
    /product
    0.07
    ful
    0.07
    -bodied
    0.07
    full
    0.07
    Activations Density 0.018%

    No Known Activations