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. 23159
    Prev
    Next
    INDEX
    Explanations

    references to actors and their conditions related to performance and comfort

    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
    WithData
    -0.18
     Wrapper
    -0.17
    WithValue
    -0.17
    WidgetItem
    -0.16
    liÄį
    -0.15
    .MockMvc
    -0.14
    Worksheet
    -0.14
    achs
    -0.14
    ylko
    -0.14
    chine
    -0.14
    POSITIVE LOGITS
    -w
    0.55
    w
    0.43
    W
    0.42
     wear
    0.40
    _w
    0.39
     wearing
    0.36
     wears
    0.35
     W
    0.34
    .w
    0.33
     wore
    0.31
    Activations Density 0.022%

    No Known Activations