© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    5. Llama3.1-8B (Base)
    7. Residual Stream
    9. 9-LLAMASCOPE-RES-131K
    11. 94237
    Prev
    Next
    INDEX
    Explanations

    instances of the word "heavily" and its related forms, indicating emphasis or intensity

    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
    erce
    -0.08
    rike
    -0.08
    ÃŃrk
    -0.07
    EMPLARY
    -0.07
     geg
    -0.07
    ~":"
    -0.07
     BITTE
    -0.07
    #
    -0.07
    seau
    -0.07
    anness
    -0.07
    POSITIVE LOGITS
    inar
    0.06
    ıma
    0.06
    ãĥ¼ãĥ¬
    0.06
    acha
    0.06
    _intent
    0.06
    ider
    0.06
    mür
    0.06
    åĩºæĿ¥
    0.06
     Farr
    0.06
     pam
    0.06
    Activations Density 0.008%

    No Known Activations