© 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. 17-LLAMASCOPE-RES-131K
    11. 129042
    Prev
    Next
    INDEX
    Explanations

    punctuation marks associated with dialogue, particularly quotation marks and exclamations

    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
    edenÃŃ
    -0.07
    .den
    -0.07
    hpp
    -0.07
    çĿĽ
    -0.07
     Seal
    -0.07
    rnd
    -0.06
    _isr
    -0.06
    opes
    -0.06
     пенÑģ
    -0.06
    pun
    -0.06
    POSITIVE LOGITS
     etc
    0.09
    etc
    0.07
     according
    0.06
    lem
    0.06
    outers
    0.06
    Ñĥла
    0.06
     crack
    0.06
    logan
    0.06
     they
    0.06
     she
    0.06
    Activations Density 0.021%

    No Known Activations