© 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. 24327
    Prev
    Next
    INDEX
    Explanations

    structured data related to certifications and training programs

    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
    -0.22
     (
    -0.21
     and
    -0.20
     re
    -0.19
     all
    -0.18
     the
    -0.18
     in
    -0.18
     for
    -0.18
     at
    -0.17
     g
    -0.17
    POSITIVE LOGITS
    .scalablytyped
    0.22
    istrovstvÃŃ
    0.18
    notes
    0.18
    actionDate
    0.18
     Description
    0.18
    styleType
    0.17
    chwitz
    0.17
    fillType
    0.17
    yonel
    0.17
    UsageId
    0.17
    Activations Density 0.088%

    No Known Activations