© 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. 21588
    Prev
    Next
    INDEX
    Explanations

    expressions and symbols related to mathematical equations and parameters

    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
    fragistics
    -0.14
    pard
    -0.14
    .WinForms
    -0.13
    illez
    -0.13
    AdapterManager
    -0.13
    urette
    -0.13
    algo
    -0.13
    -Semit
    -0.12
    efe
    -0.12
    ADOR
    -0.12
    POSITIVE LOGITS
    âĢª
    0.14
     tiên
    0.13
    _BINDING
    0.12
    âĢ¦↵
    0.12
    лÑĸÑĤ
    0.12
    ">ÃĹ</
    0.12
    ÌĢ
    0.12
     createAction
    0.12
     [âĢ¦]↵
    0.12
    ÌĤ
    0.12
    Activations Density 0.002%

    No Known Activations