© 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
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 63291
    Prev
    Next
    INDEX
    Explanations

    references to publications and citations in academic writing

    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
    pad
    -0.17
    ye
    -0.16
    XA
    -0.15
    itz
    -0.14
    apid
    -0.14
    uzz
    -0.14
    ä¸Ģåį·
    -0.13
    arel
    -0.13
     kidding
    -0.13
    ibr
    -0.13
    POSITIVE LOGITS
     pass
    0.20
     p
    0.18
     pp
    0.18
     especially
    0.17
    omap
    0.15
     esp
    0.15
    ẫ
    0.15
     see
    0.14
    žen
    0.14
    seealso
    0.14
    Activations Density 0.044%

    No Known Activations