© 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. 22356
    Prev
    Next
    INDEX
    Explanations

    references to "end" or the conclusion of events or topics

    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
    iously
    -0.14
    raith
    -0.14
    ambique
    -0.14
    seys
    -0.14
    à¹ĥà¸Ī
    -0.14
    ì§Ī
    -0.14
    ãģ¾ãģ¾
    -0.14
    jerne
    -0.13
    otate
    -0.13
    ivot
    -0.13
    POSITIVE LOGITS
     of
    0.25
    game
    0.24
    /start
    0.21
    orse
    0.19
    earing
    0.19
    ereço
    0.18
     Credits
    0.18
    owment
    0.18
    ogenous
    0.17
     credits
    0.17
    Activations Density 0.029%

    No Known Activations