© 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. 15-LLAMASCOPE-RES-131K
    11. 417
    Prev
    Next
    INDEX
    Explanations

    references to messiness or dirtiness

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
    ä¼
    -0.06
    ABCDE
    -0.06
    arga
    -0.06
     field
    -0.06
     Foreign
    -0.06
    Inlining
    -0.05
     Mickey
    -0.05
    [âĢ¦
    -0.05
    ÑĥзÑĭ
    -0.05
     unwrap
    -0.05
    POSITIVE LOGITS
    áš
    0.08
    orer
    0.08
    utow
    0.07
    ncoder
    0.07
     {{--<
    0.07
    ordova
    0.07
    ocz
    0.07
    metatable
    0.07
    ycz
    0.06
    é»
    0.06
    Activations Density 0.013%

    No Known Activations