© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 26201
    Prev
    Next
    INDEX
    Explanations

    references to chemical compounds and their properties

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Negative Logits
     nurses
    -0.27
     Nurses
    -0.25
     nails
    -0.25
     nitrogen
    -0.24
     noodles
    -0.23
     nose
    -0.22
     nursing
    -0.22
     Nursing
    -0.22
    _NOTIFICATION
    -0.21
     notification
    -0.20
    POSITIVE LOGITS
    prenom
    0.16
     gross
    0.16
    Surname
    0.15
    zier
    0.15
     surname
    0.14
    -positive
    0.14
    MBED
    0.14
    ìľł
    0.14
    ieren
    0.14
     Julio
    0.14
    Activations Density 0.023%

    No Known Activations