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. 43887
    Prev
    Next
    INDEX
    Explanations

    expressions of negation or uncertainty

    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
     EITHER
    -0.09
    _UNICODE
    -0.07
    (åľŁ
    -0.07
     Neither
    -0.07
    azor
    -0.07
     SUCH
    -0.07
    аннÑı
    -0.07
    .big
    -0.06
    cles
    -0.06
     neither
    -0.06
    POSITIVE LOGITS
     downright
    0.19
     outright
    0.17
     actually
    0.12
     border
    0.12
     flat
    0.11
     actively
    0.11
     verg
    0.11
     occasionally
    0.10
     altogether
    0.10
     borderline
    0.10
    Activations Density 0.042%

    No Known Activations