© 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. 70368
    Prev
    Next
    INDEX
    Explanations

    phrases that include "and" or "or" to indicate conjunctions or options

    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
    ovit
    -0.08
    acci
    -0.07
    ät
    -0.07
    pais
    -0.07
    peria
    -0.07
    llib
    -0.07
    lamaz
    -0.07
    ioned
    -0.07
    eted
    -0.07
    aklı
    -0.07
    POSITIVE LOGITS
    or
    0.08
     especially
    0.07
    AMP
    0.06
    ye
    0.06
    459
    0.06
     tent
    0.05
     meiden
    0.05
    ines
    0.05
    esp
    0.05
    /or
    0.05
    Activations Density 0.096%

    No Known Activations