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. 23695
    Prev
    Next
    INDEX
    Explanations

    references to breeding, particularly in the context of societal issues or debates

    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
    relude
    -0.19
    ardu
    -0.15
    bidden
    -0.15
    à¥Ģà¤ķ
    -0.14
    untu
    -0.14
    aneous
    -0.14
    eriod
    -0.14
     jejichž
    -0.14
    oooooooo
    -0.14
    SON
    -0.14
    POSITIVE LOGITS
    htaking
    0.22
    neck
    0.20
    ers
    0.19
    emer
    0.17
    enden
    0.17
    ings
    0.17
    leur
    0.17
    jamin
    0.17
    weet
    0.16
    winner
    0.16
    Activations Density 0.024%

    No Known Activations