© 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
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 30603
    Prev
    Next
    INDEX
    Explanations

    phrases related to the act of removing or changing something

    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
     wounds
    -0.15
    ycop
    -0.14
    reme
    -0.14
    strap
    -0.14
    sla
    -0.14
    й
    -0.14
    IRMWARE
    -0.13
    fp
    -0.13
    cz
    -0.13
    loss
    -0.13
    POSITIVE LOGITS
     khá»ıi
    0.22
     from
    0.19
    /disable
    0.19
     obstacles
    0.18
     imped
    0.17
     barriers
    0.17
    /add
    0.17
    _from
    0.17
    ">ÃĹ</
    0.17
     tạp
    0.17
    Activations Density 0.065%

    No Known Activations