Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 23636
    Prev
    Next
    INDEX
    Explanations

    the beginning of a document or section

    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
    ://
    -0.21
    _REF
    -0.17
    .Forms
    -0.17
    php
    -0.16
     Angeles
    -0.16
    .swing
    -0.13
     Kong
    -0.13
    .S
    -0.12
    ysics
    -0.11
    ="
    -0.11
    POSITIVE LOGITS
    .GraphicsUnit
    0.10
    #echo
    0.10
    ë¨
    0.09
    podob
    0.09
    aeda
    0.09
    اسطة
    0.09
    ê¸Ķ
    0.09
    ÙĪÛĮÙĨت
    0.09
    (íģ¬ê¸°
    0.09
    ÙĪÛĮزÛĮ
    0.08
    Activations Density 0.800%

    No Known Activations