© 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 EvalsExportsinterp-engineNEWAPI 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. 12936
    Prev
    Next
    INDEX
    Explanations

    descriptions related to locations and landmarks

    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
    ÐĴаж
    -0.19
     ÐĶжон
    -0.16
     Чи
    -0.16
     ÐļÑĢи
    -0.16
    оÑģÑĥд
    -0.16
    -Saharan
    -0.15
     коÑĢол
    -0.15
     ÐĿÑĸ
    -0.14
     индивидÑĥ
    -0.14
     Aydın
    -0.14
    POSITIVE LOGITS
     Moscow
    0.40
     Russia
    0.37
     Russian
    0.36
     Siber
    0.32
     Russians
    0.32
     Petersburg
    0.32
    Russian
    0.31
    Russia
    0.30
     Soviet
    0.30
     Putin
    0.28
    Activations Density 0.071%

    No Known Activations