© 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. 25292
    Prev
    Next
    INDEX
    Explanations

    references to geographical locations or significant entities associated with places

    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
    aea
    -0.15
    ç¶
    -0.14
    353
    -0.13
    ERING
    -0.13
    ¿ł
    -0.13
    707
    -0.13
    <::
    -0.13
    incr
    -0.13
     Wikip
    -0.13
    tright
    -0.13
    POSITIVE LOGITS
     join
    1.11
     joins
    1.05
     joining
    1.04
     Join
    1.03
    join
    0.97
    Join
    0.94
     joined
    0.90
    joining
    0.87
     JOIN
    0.84
    .join
    0.81
    Activations Density 0.013%

    No Known Activations