© 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 EvalsExportsinterp-engineNEWAPI 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. 94952
    Prev
    Next
    INDEX
    Explanations

    programming-related keywords and error handling terms

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

    Method or property access in programming code, often within program structures, where the token contains significant function names, class identifiers, or property access syntax.

    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
    Ñİк
    -0.07
    ly
    -0.07
    ed
    -0.07
    .Reporting
    -0.07
    esion
    -0.07
    .MixedReality
    -0.06
    uite
    -0.06
    LLU
    -0.06
    ãĤ§
    -0.06
     jež
    -0.06
    POSITIVE LOGITS
    IGINAL
    0.10
    crease
    0.10
    ftware
    0.10
    avy
    0.10
    egral
    0.09
    bsite
    0.09
    ategy
    0.09
    enticate
    0.09
    ergency
    0.09
    ctrine
    0.09
    Activations Density 0.270%

    No Known Activations