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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 76484
    Prev
    Next
    INDEX
    Explanations

    dialogue and conversational elements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Quotation marks and accompanying punctuation (periods, commas, question marks) that indicate the beginning or end of direct speech, often paired with dialogue attribution.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    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
    meli
    -0.07
    楽ãģĹ
    -0.07
    ä¸įå¾Ĺ
    -0.06
    ptest
    -0.06
    ungal
    -0.06
    igo
    -0.06
     Classics
    -0.06
    710
    -0.06
    erre
    -0.06
    _ir
    -0.06
    POSITIVE LOGITS
     sure
    0.08
     YES
    0.08
     yes
    0.08
    lef
    0.07
    linger
    0.07
     LIS
    0.07
     fine
    0.07
    yes
    0.07
     Yes
    0.06
    lete
    0.06
    Activations Density 0.043%

    No Known Activations