© 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. 104322
    Prev
    Next
    INDEX
    Explanations

    references to the UAE (United Arab Emirates) and its significance in trade and relationships

    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
    arih
    -0.08
    letics
    -0.07
    anel
    -0.07
    oir
    -0.07
    nette
    -0.07
    ught
    -0.07
    aran
    -0.07
    epad
    -0.07
    activ
    -0.07
    atics
    -0.07
    POSITIVE LOGITS
    -wide
    0.10
    -Israel
    0.09
    -based
    0.07
    anness
    0.07
    tember
    0.07
     nationals
    0.07
    s
    0.06
    -region
    0.06
    Å¡ti
    0.06
    -origin
    0.06
    Activations Density 0.004%

    No Known Activations