© 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. 33410
    Prev
    Next
    INDEX
    Explanations

    references to collaboration or assistance

    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
     retros
    -0.18
     Restoration
    -0.18
     Rat
    -0.17
    Retention
    -0.16
     Roger
    -0.16
    rana
    -0.16
    restore
    -0.16
    Roger
    -0.15
    rollo
    -0.15
     retrospective
    -0.15
    POSITIVE LOGITS
    -re
    0.80
     RE
    0.77
     Reed
    0.77
    -Re
    0.77
    Re
    0.74
     Reynolds
    0.69
     Reid
    0.69
    re
    0.67
    (re
    0.65
    	re
    0.65
    Activations Density 0.011%

    No Known Activations