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. 44086
    Prev
    Next
    INDEX
    Explanations

    references to impurities 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
     prospect
    -0.07
    ABEL
    -0.07
    ABLE
    -0.06
    豪
    -0.06
    åĹ
    -0.06
    742
    -0.06
    (es
    -0.06
    над
    -0.06
    atorium
    -0.06
    sian
    -0.06
    POSITIVE LOGITS
    urities
    0.24
    urity
    0.20
    ure
    0.14
     purity
    0.12
     pur
    0.11
    URITY
    0.11
    regn
    0.10
     Pur
    0.10
    unities
    0.09
    urer
    0.09
    Activations Density 0.007%

    No Known Activations