© 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. 16-LLAMASCOPE-RES-32K
    11. 17923
    Prev
    Next
    INDEX
    Explanations

    terms associated with workplace policies and structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k
    Configuration
    fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    cerebras/SlimPajama-627B
    Features
    32,768
    Data Type
    bfloat16
    Hook Name
    blocks.16.hook_resid_post
    Hook Layer
    16
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    cum
    -0.14
    VERR
    -0.14
    \grid
    -0.13
     ìŀĪëĬĶëį°
    -0.13
    -and
    -0.13
    VELO
    -0.13
    ulf
    -0.13
    DataReader
    -0.13
    afa
    -0.13
    aban
    -0.13
    POSITIVE LOGITS
    ï¼Į以åıĬ
    0.39
     oraz
    0.31
    以åıĬ
    0.27
     sowie
    0.26
     plus
    0.26
     samt
    0.25
     ê·¸ë¦¬ê³ł
    0.22
     ainsi
    0.21
    ;
    0.20
     PLUS
    0.19
    Activations Density 0.883%

    No Known Activations