© 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. 31-LLAMASCOPE-RES-32K
    11. 14704
    Prev
    Next
    INDEX
    Explanations

    references to energy efficiency in buildings and the associated technologies

    oai_token-act-pair · gpt-4o-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
    cus
    -0.15
    abus
    -0.15
     cá»Ń
    -0.14
    ^K
    -0.14
    RunLoop
    -0.13
    )prepare
    -0.13
    ãĤ¯ãĤ»
    -0.13
    agos
    -0.12
    lio
    -0.12
    emade
    -0.12
    POSITIVE LOGITS
     Passive
    0.40
     passive
    0.33
     Pass
    0.32
     thermal
    0.31
    Pass
    0.28
     retrofit
    0.28
     buildings
    0.27
     Thermal
    0.27
     envelope
    0.26
    thermal
    0.26
    Activations Density 0.048%

    No Known Activations