© 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. 44166
    Prev
    Next
    INDEX
    Explanations

    indications of lengthy legal documents or detailed reports

    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
    cie
    -0.16
    ivals
    -0.16
    owie
    -0.14
    iets
    -0.14
     Paren
    -0.14
    -desc
    -0.14
     Summers
    -0.14
    oro
    -0.14
    iece
    -0.13
    ipers
    -0.13
    POSITIVE LOGITS
     pages
    0.51
     page
    0.47
    -page
    0.40
     Pages
    0.38
    pages
    0.37
     Page
    0.37
    Page
    0.36
    page
    0.36
    Pages
    0.36
    -pages
    0.35
    Activations Density 0.051%

    No Known Activations