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

    Neuronpedia

    Jacobian LensNEW
    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. 43192
    Prev
    Next
    INDEX
    Explanations

    parts of the document related to significant events or announcements

    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
    à¹Ĥย
    -0.13
    pyx
    -0.12
     Rodriguez
    -0.12
    ideshow
    -0.12
     unidentified
    -0.12
    vier
    -0.12
    è¯ij
    -0.12
    istor
    -0.12
    bbe
    -0.12
     TBD
    -0.11
    POSITIVE LOGITS
    adora
    0.14
    ezi
    0.13
    onical
    0.13
     discrepan
    0.13
    icari
    0.13
    ัล
    0.13
    кÑĸн
    0.13
    354
    0.13
    aturated
    0.12
    amburg
    0.12
    Activations Density 0.368%

    No Known Activations