© 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. 17998
    Prev
    Next
    INDEX
    Explanations

    interrogative and declarative statements, particularly focusing on instructions and questions

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

    The token ":" typically appears after the letter "A" at the beginning of an answer or response section in technical forums, marking the transition from a question to an answer in programming/technical Q&A contexts.

    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
    丸
    -0.07
    itched
    -0.07
    AGO
    -0.07
    oux
    -0.07
    deen
    -0.07
    ardown
    -0.07
    vale
    -0.07
    ä¸įå¾Ĺ
    -0.07
    ãĤ¤ãĥ³ãĥĪ
    -0.07
    sert
    -0.07
    POSITIVE LOGITS
     easiest
    0.08
     simply
    0.07
     try
    0.07
     you
    0.07
     use
    0.07
     Use
    0.07
     simplest
    0.07
     solution
    0.06
    otron
    0.06
    Use
    0.06
    Activations Density 0.008%

    No Known Activations