© 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
    2. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    3. Llama3.1-8B (Base)
    4. Residual Stream
    5. 17-LLAMASCOPE-RES-131K
    6. 130237
    Prev
    Next
    INDEX
    Explanations

    guidelines or instructions related to data management and formatting tasks

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

    Highlighted tokens are often common nouns or noun phrases that represent specific concepts, physical objects, or time periods relevant to the surrounding text. They frequently appear at syntactically important positions where they serve as subjects, objects, or modifiers that ground the meaning of the surrounding content.

    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
    ifton
    -0.07
     sice
    -0.07
    thinkable
    -0.07
    .setViewport
    -0.07
    ruba
    -0.07
    znik
    -0.07
    à¥Ģमत
    -0.07
    ylon
    -0.06
    tero
    -0.06
    ảo
    -0.06
    POSITIVE LOGITS
    isque
    0.07
    éĢģæĸĻçĦ¡æĸĻ
    0.07
    qué
    0.07
     dating
    0.06
    ãĢģ 
    0.06
    uncios
    0.06
    istrovstvÃŃ
    0.06
     Sidd
    0.05
    iores
    0.05
    quot
    0.05
    Activations Density 1.332%

    No Known Activations