© 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. 13002
    Prev
    Next
    INDEX
    Explanations

    numeric data and financial figures

    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
     ÙħÛĮÙĦادÛĮ
    -0.13
    OrNull
    -0.13
    /off
    -0.12
    buat
    -0.12
    ree
    -0.12
    âĢļ
    -0.11
    ông
    -0.11
    cho
    -0.11
     Boeh
    -0.11
    reater
    -0.11
    POSITIVE LOGITS
    ivery
    0.14
    togroup
    0.14
    ubyte
    0.13
    Ľå»º
    0.13
    igure
    0.13
    itelist
    0.13
    моÑĤ
    0.13
    arsimp
    0.12
    ibox
    0.12
    °ëĭ¤
    0.12
    Activations Density 0.185%

    No Known Activations