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. 23724
    Prev
    Next
    INDEX
    Explanations

    references to police activity and response

    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
    ĩ
    -0.15
     expiresIn
    -0.15
    .Sync
    -0.15
    estro
    -0.14
    ivil
    -0.14
    ág
    -0.14
    å§Ķåijĺä¼ļ
    -0.14
     numberWith
    -0.14
    azo
    -0.14
    á¾
    -0.14
    POSITIVE LOGITS
     summoned
    0.18
     called
    0.17
     alerted
    0.17
     attendance
    0.17
     receipt
    0.17
     receive
    0.17
     received
    0.16
     Called
    0.16
     responded
    0.16
     contacted
    0.16
    Activations Density 0.014%

    No Known Activations