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. 23580
    Prev
    Next
    INDEX
    Explanations

    references to politicians and their statements or positions

    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
    agers
    -0.16
    unda
    -0.15
     Legislative
    -0.15
     council
    -0.14
    /operators
    -0.14
     legislative
    -0.14
    ERRU
    -0.14
    oi
    -0.14
    rez
    -0.14
     å·Ŀ
    -0.14
    POSITIVE LOGITS
     Shadow
    0.28
     shadow
    0.27
    Shadow
    0.24
    shadow
    0.23
    -shadow
    0.20
     Health
    0.20
     Communities
    0.19
     Culture
    0.19
     Transport
    0.18
     Home
    0.18
    Activations Density 0.057%

    No Known Activations