© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 14253
    Prev
    Next
    INDEX
    Explanations

    references to demographic changes and geographic regions within the UK

    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
    elia
    -0.15
    anut
    -0.15
    ãĤ¿ãĤ¤
    -0.15
    æĭĽ
    -0.14
    essor
    -0.14
    ijn
    -0.14
     tuto
    -0.14
     Continental
    -0.14
    rett
    -0.14
     Fra
    -0.14
    POSITIVE LOGITS
     UK
    0.35
    UK
    0.28
     uk
    0.25
    .uk
    0.25
     NHS
    0.21
     Uk
    0.20
    addir
    0.19
     country
    0.19
     Channel
    0.19
     Duch
    0.19
    Activations Density 0.052%

    No Known Activations