© 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. 25756
    Prev
    Next
    INDEX
    Explanations

    concepts related to data protection and usage in programming contexts

    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
    IFI
    -0.13
     èĭ¥
    -0.13
    Ascii
    -0.13
    /UIKit
    -0.13
     torture
    -0.13
    _MACRO
    -0.12
    ire
    -0.12
    кÑĥлÑı
    -0.12
    à¹ĩà¸Ķ
    -0.12
    umps
    -0.12
    POSITIVE LOGITS
    ramer
    0.15
    aines
    0.15
     concept
    0.14
     dealing
    0.14
    igans
    0.14
    ynchronous
    0.14
     dealt
    0.14
    kova
    0.13
    enstein
    0.13
     {};č↵
    0.13
    Activations Density 0.422%

    No Known Activations