© 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. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 43924
    Prev
    Next
    INDEX
    Explanations

    scientific terminology related to biological processes and their consequences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Juliushanhanhan/llama-3-8b-it-res/blocks.25.hook_resid_post
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    gated
    Context Size
    1,024
    Dataset
    Juliushanhanhan/openwebtext-1b-llama3-tokenized-cxt-1024
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     accompl
    -0.14
    omi
    -0.14
    regor
    -0.12
    å¸Ń
    -0.12
    amage
    -0.12
     literally
    -0.12
     wound
    -0.12
     punish
    -0.12
    .pag
    -0.12
     comple
    -0.12
    POSITIVE LOGITS
     lead
    0.29
     leading
    0.25
     result
    0.25
     resulted
    0.25
     leads
    0.24
     cause
    0.23
     led
    0.23
    result
    0.22
     may
    0.21
     Result
    0.19
    Activations Density 0.373%

    No Known Activations