© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 40727
    Prev
    Next
    INDEX
    Explanations

    references to food and technology components

    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
    _Tis
    -0.20
    $MESS
    -0.18
    $LANG
    -0.17
    /loose
    -0.17
    >NN
    -0.17
    _Lean
    -0.16
     cryst
    -0.16
    _regeneration
    -0.15
    /Gate
    -0.15
     reelection
    -0.15
    POSITIVE LOGITS
    -
    0.33
     -
    0.22
    \_
    0.18
    -style
    0.16
    Âł
    0.16
    -level
    0.16
    -wide
    0.16
    -type
    0.15
    -looking
    0.15
    imap
    0.15
    Activations Density 0.080%

    No Known Activations