© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    5. Llama3.1-8B (Base)
    7. Residual Stream
    9. 22-LLAMASCOPE-RES-131K
    11. 4237
    Prev
    Next
    INDEX
    Explanations

    references to sports and public figures, particularly in the context of announcements and reactions

    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
    Äįas
    -0.08
     bazen
    -0.07
    ään
    -0.07
    ÙĪÙĨا
    -0.07
    hlas
    -0.07
    opup
    -0.07
    ůj
    -0.07
    ãĥ³ãĥĶ
    -0.07
     arada
    -0.07
    ETCH
    -0.07
    POSITIVE LOGITS
     reveal
    0.07
     something
    0.07
     secret
    0.07
     completely
    0.07
     exactly
    0.07
    .
    0.07
     This
    0.07
     reveals
    0.06
     this
    0.06
    ioned
    0.06
    Activations Density 0.018%

    No Known Activations