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. 31359
    Prev
    Next
    INDEX
    Explanations

    references to the general public and societal norms

    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
    â̦
    -0.14
     unto
    -0.14
     particular
    -0.13
    Ì
    -0.13
    ...↵
    -0.13
    â̦↵
    -0.13
     specific
    -0.13
     entire
    -0.12
    ych
    -0.12
    â̦the
    -0.12
    POSITIVE LOGITS
    /original
    0.24
    liest
    0.18
    oard
    0.16
     ########.
    0.16
    zers
    0.15
    ฯ
    0.15
    est
    0.15
    iest
    0.15
    aviest
    0.15
    /current
    0.15
    Activations Density 1.053%

    No Known Activations