© 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. 21167
    Prev
    Next
    INDEX
    Explanations

    references to specific chapters or sections in academic or analytical texts

    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
    _EOL
    -0.16
    #
    -0.15
    ewe
    -0.15
    ýn
    -0.14
    EditingController
    -0.14
    á»ĩu
    -0.14
    º
    -0.14
    ÏĢη
    -0.14
    :href
    -0.13
    ongyang
    -0.13
    POSITIVE LOGITS
     chapter
    0.97
     Chapter
    0.85
     chapters
    0.83
    chapter
    0.76
    Chapter
    0.75
     Chapters
    0.67
     section
    0.62
     CHAPTER
    0.62
    ç«ł
    0.61
    .chapter
    0.57
    Activations Density 0.024%

    No Known Activations