© 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. 94863
    Prev
    Next
    INDEX
    Explanations

    proper nouns and their associations in storytelling contexts

    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
    Called
    -0.09
     Called
    -0.09
     titled
    -0.09
     called
    -0.09
    called
    -0.08
    aeda
    -0.08
    itled
    -0.08
    ucher
    -0.07
    ceae
    -0.07
    gae
    -0.07
    POSITIVE LOGITS
     whose
    0.07
     who
    0.07
     which
    0.07
    nde
    0.07
     whom
    0.07
     most
    0.07
     cui
    0.06
    efs
    0.06
    )
    0.06
     engineered
    0.06
    Activations Density 0.041%

    No Known Activations