© 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
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 15453
    Prev
    Next
    INDEX
    Explanations

    instances of the word "two" in various contexts

    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
     Thousands
    -0.20
     thousands
    -0.17
    Thousands
    -0.15
    sian
    -0.15
    stå
    -0.15
    aira
    -0.15
    amen
    -0.14
    åIJĦç§į
    -0.14
    stakes
    -0.14
    .getElements
    -0.14
    POSITIVE LOGITS
     separate
    0.30
    ãģ¤ãģ®
    0.25
    fold
    0.25
     different
    0.25
     consecutive
    0.24
     distinct
    0.24
     or
    0.23
     sets
    0.23
    -thirds
    0.23
    ãģ¤
    0.21
    Activations Density 0.289%

    No Known Activations