© 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. 8438
    Prev
    Next
    INDEX
    Explanations

    references to specific high schools and their sports teams

    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
    fdc
    -0.17
     Sundays
    -0.15
     lecturer
    -0.15
    LECT
    -0.15
     lectures
    -0.15
     nursery
    -0.14
    pio
    -0.14
     undergrad
    -0.14
     rookies
    -0.14
     UFC
    -0.14
    POSITIVE LOGITS
     high
    0.31
    HS
    0.31
     vars
    0.30
     High
    0.30
     RHS
    0.29
     NHS
    0.26
    High
    0.26
    vars
    0.24
    high
    0.24
     HS
    0.24
    Activations Density 0.074%

    No Known Activations