© 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. 31-LLAMASCOPE-RES-32K
    11. 22203
    Prev
    Next
    INDEX
    Explanations

    descriptive elements related to architectural features and their historical significance

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    iddet
    -0.14
    prestashop
    -0.13
    SGlobal
    -0.13
    APTER
    -0.13
     Antar
    -0.13
    vyk
    -0.13
    ruba
    -0.13
    hq
    -0.13
    celik
    -0.13
    osti
    -0.13
    POSITIVE LOGITS
     Formal
    0.29
     formal
    0.29
     park
    0.26
     arb
    0.25
     ponds
    0.25
     landsc
    0.25
     law
    0.24
     lawn
    0.24
     gardens
    0.23
     manic
    0.23
    Activations Density 0.101%

    No Known Activations