© 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. 21-LLAMASCOPE-RES-32K
    11. 4707
    Prev
    Next
    INDEX
    Explanations

    measurements and descriptions related to geometry and positioning

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Negative Logits
    iset
    -0.15
    462
    -0.15
     segment
    -0.15
    .baidu
    -0.14
    AZY
    -0.14
    byt
    -0.14
    antar
    -0.14
    estic
    -0.14
    entials
    -0.14
    stras
    -0.14
    POSITIVE LOGITS
     nearest
    0.20
     edge
    0.20
     center
    0.20
    axis
    0.20
     ÑĨенÑĤÑĢа
    0.19
     axis
    0.18
     centre
    0.18
    леж
    0.18
    axes
    0.18
    edge
    0.17
    Activations Density 0.215%

    No Known Activations