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

    Neuronpedia

    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. 127715
    Prev
    Next
    INDEX
    Explanations

    references to coding or markdown elements

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

    Special characters and syntax elements in programming code, including brackets, operators, HTML/XML tags, and programming language specific keywords that form valid syntax in languages like JavaScript, CSS, PHP, Java, and markup languages.

    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
    atis
    -0.07
    ÏĢοÏĦε
    -0.06
    ateur
    -0.06
    grim
    -0.06
    zin
    -0.06
    taboola
    -0.06
    æŁĦ
    -0.06
    crest
    -0.06
    ajo
    -0.06
    arters
    -0.06
    POSITIVE LOGITS
    blr
    0.08
     fenced
    0.07
     embed
    0.07
    oger
    0.06
    .Void
    0.06
     Bul
    0.06
    åµ
    0.06
    hti
    0.06
    osci
    0.06
     bul
    0.06
    Activations Density 0.021%

    No Known Activations