© 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. 21-LLAMASCOPE-RES-32K
    11. 19277
    Prev
    Next
    INDEX
    Explanations

    legal and religious terminology related to marriage and canonical law

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    tics
    -0.15
     enn
    -0.14
    Ñijм
    -0.14
     trom
    -0.14
    Outlined
    -0.14
     Bian
    -0.14
    çŃ
    -0.14
    ikal
    -0.14
     en
    -0.14
    reon
    -0.14
    POSITIVE LOGITS
    orum
    0.22
     Laurent
    0.19
     quam
    0.19
     reddit
    0.17
    culus
    0.17
     quo
    0.17
     Sty
    0.16
    rum
    0.16
    actus
    0.16
     redd
    0.16
    Activations Density 0.071%

    No Known Activations