© 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. 29655
    Prev
    Next
    INDEX
    Explanations

    terms related to variants and classifications of products or concepts

    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
     <<-
    -0.15
     ([[
    -0.15
    ì±
    -0.15
    .↵↵↵↵↵↵↵↵
    -0.14
    /**č↵
    -0.14
    ëĭ
    -0.14
     [âĢ¦]...↵
    -0.14
    ENCH
    -0.13
    .↵↵↵↵↵↵↵↵↵↵↵↵
    -0.13
    виÑĤ
    -0.13
    POSITIVE LOGITS
     Edit
    0.42
    Edit
    0.40
     [
    0.38
    edit
    0.38
     edit
    0.36
     ç¼ĸè¾ij
    0.32
    _[
    0.29
    -edit
    0.29
    _edit
    0.28
    [
    0.24
    Activations Density 0.073%

    No Known Activations