© 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. 15031
    Prev
    Next
    INDEX
    Explanations

    key terms and phrases related to authority, governance, and social structures

    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
    /from
    -0.34
    /of
    -0.25
    /to
    -0.23
    /on
    -0.21
    /her
    -0.21
    /out
    -0.19
    /by
    -0.19
    /or
    -0.19
     ниÑħ
    -0.18
    /as
    -0.17
    POSITIVE LOGITS
    ä¸Ģä¸ĭ
    0.28
    çļĦæĺ¯
    0.22
    (ed
    0.19
    ively
    0.18
    ing
    0.17
    /report
    0.16
    ulate
    0.16
    ned
    0.16
    ize
    0.16
    çļĦäºĭæĥħ
    0.16
    Activations Density 2.905%

    No Known Activations