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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 29155
    Prev
    Next
    INDEX
    Explanations

    references to stages and phases within structured processes or analyses

    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
    à¸ļาล
    -0.17
    erken
    -0.15
     ÄĮer
    -0.15
    YRO
    -0.15
     latter
    -0.14
    /ros
    -0.14
    .au
    -0.14
    rý
    -0.13
    ì²Ļ
    -0.13
     Ñģобой
    -0.13
    POSITIVE LOGITS
     of
    0.23
    (s
    0.20
     (
    0.18
    /single
    0.18
    缮ãģ®
    0.17
     onwards
    0.17
    -third
    0.17
     của
    0.16
    à¸Ĥà¸Ńà¸ĩ
    0.16
    isode
    0.16
    Activations Density 0.099%

    No Known Activations