© 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 EvalsExportsAPI 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. 58840
    Prev
    Next
    INDEX
    Explanations

    proper nouns related to names and places, particularly those containing the substring "har" or "Bhar."

    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
    ForObject
    -0.09
    s
    -0.08
    meer
    -0.07
    ιά
    -0.07
    olls
    -0.07
    pressions
    -0.07
    Ñĩини
    -0.07
    avors
    -0.06
    uvre
    -0.06
    é¢Ŀ
    -0.06
    POSITIVE LOGITS
     Kre
    0.07
     Pf
    0.06
    utut
    0.06
    anter
    0.06
    hardt
    0.06
     Imperial
    0.06
    å¾Ĵ
    0.06
    bette
    0.06
     Schwarz
    0.06
    foreground
    0.06
    Activations Density 0.003%

    No Known Activations