© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 22911
    Prev
    Next
    INDEX
    Explanations

    statistical data related to abortion rates and gestational weeks

    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
    رÛĮز
    -0.17
     Pricing
    -0.13
    št
    -0.13
    ertz
    -0.13
    ENO
    -0.13
    аÑĢам
    -0.13
    Multiply
    -0.12
    uyết
    -0.12
     {{{
    -0.12
    Mill
    -0.12
    POSITIVE LOGITS
     share
    0.50
     proportion
    0.42
    share
    0.39
     number
    0.39
     percentage
    0.37
     fraction
    0.35
     Share
    0.35
    Share
    0.33
    -share
    0.32
    _share
    0.32
    Activations Density 0.188%

    No Known Activations