© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
    5. Llama3.1-8B (Base)
    7. Residual Stream
    9. 20-LLAMASCOPE-RES-32K
    11. 32529
    Prev
    Next
    INDEX
    Explanations

    references to publication details, including issues, articles, and journals

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    yan
    -0.14
     gre
    -0.14
     Ans
    -0.14
    iete
    -0.14
     OSD
    -0.14
    æĻ´
    -0.13
    æ³
    -0.13
    cores
    -0.13
    LEC
    -0.13
    .desktop
    -0.13
    POSITIVE LOGITS
     issue
    0.33
     edition
    0.31
     episode
    0.30
     latest
    0.29
     Issue
    0.27
    æľĢæĸ°
    0.26
    episode
    0.26
    issue
    0.25
     editions
    0.25
    /latest
    0.25
    Activations Density 0.099%

    No Known Activations