© 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. 24-LLAMASCOPE-RES-32K
    11. 10354
    Prev
    Next
    INDEX
    Explanations

    elements related to film production and acting credits

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
     flown
    -0.17
    ãģ¨ãģª
    -0.17
     bitten
    -0.17
     beaten
    -0.17
     spoken
    -0.16
    oken
    -0.16
     shaken
    -0.16
    dden
    -0.16
    .Companion
    -0.15
     risen
    -0.15
    POSITIVE LOGITS
     hizo
    0.37
     pudo
    0.32
    uvo
    0.31
     baktı
    0.31
     yaptı
    0.31
    rió
    0.30
     was
    0.30
     oldu
    0.29
     took
    0.29
     fue
    0.28
    Activations Density 0.085%

    No Known Activations