© 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. 81774
    Prev
    Next
    INDEX
    Explanations

    the keyword "keep" and its variations, reflecting themes of persistence or continuity

    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
    ÑĦÑĸк
    -0.08
    oppable
    -0.07
    bor
    -0.07
    281
    -0.07
    ragon
    -0.07
    URITY
    -0.07
    aybe
    -0.07
    rema
    -0.07
    AGON
    -0.07
    ør
    -0.06
    POSITIVE LOGITS
    ers
    0.07
    pler
    0.06
    ahan
    0.06
    enco
    0.06
    ings
    0.06
    oval
    0.06
    ins
    0.06
    able
    0.06
    eden
    0.06
    heat
    0.06
    Activations Density 0.010%

    No Known Activations