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

    Neuronpedia

    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. 63600
    Prev
    Next
    INDEX
    Explanations

    statistics related to demographics and economic conditions

    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
     America
    -0.20
     cultures
    -0.20
    America
    -0.19
     American
    -0.19
    ãĤ¢ãĥ¡ãĥªãĤ«
    -0.18
     амеÑĢикан
    -0.17
     Americ
    -0.17
    ç¾İåľĭ
    -0.17
    American
    -0.17
    frican
    -0.17
    POSITIVE LOGITS
     citizens
    0.21
     parliament
    0.20
     urban
    0.18
     nationals
    0.17
     tertiary
    0.17
     registered
    0.17
     citizen
    0.17
     working
    0.17
     eligible
    0.17
     Parliament
    0.17
    Activations Density 0.036%

    No Known Activations