Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    3. Llama3.1-8B (Base)
    4. Residual Stream
    5. 31-LLAMASCOPE-RES-131K
    6. 44054
    Prev
    Next
    INDEX
    Explanations

    themes of navigation and guidance in various 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
    vil
    -0.08
    agner
    -0.07
     Äijá»Ļt
    -0.07
    ìĹ´
    -0.07
    redo
    -0.07
    ãģIJ
    -0.07
    èĭ±éĽĦ
    -0.07
     Äijá»į
    -0.06
    uyá»ĩt
    -0.06
    anking
    -0.06
    POSITIVE LOGITS
     chop
    0.13
     waters
    0.13
     through
    0.12
     toward
    0.11
     towards
    0.11
     tricky
    0.10
     terrain
    0.10
     Waters
    0.10
     unfamiliar
    0.10
     murky
    0.10
    Activations Density 0.019%

    No Known Activations