© 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. 12212
    Prev
    Next
    INDEX
    Explanations

    numerical data and software versioning information

    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
    ếp
    -0.15
    uard
    -0.15
    пиÑģ
    -0.15
    é½
    -0.15
    åĿª
    -0.14
    @student
    -0.14
     Ih
    -0.14
    èn
    -0.14
    196
    -0.13
    ợ
    -0.13
    POSITIVE LOGITS
    .x
    0.22
    .X
    0.18
    -Clause
    0.15
    -го
    0.15
    .latest
    0.15
     dotted
    0.15
    -transitional
    0.14
    brook
    0.14
     Harden
    0.14
    ê³Ħ
    0.13
    Activations Density 0.028%

    No Known Activations