© 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 EvalsExportsinterp-engineNEWAPI 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. 22599
    Prev
    Next
    INDEX
    Explanations

    discussions about LGBT rights and related political opinions

    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
     Barack
    -0.21
    Obama
    -0.20
     Obama
    -0.17
     Romney
    -0.17
     Occupy
    -0.17
    ActionBar
    -0.17
    bil
    -0.16
    ì¦Ŀ
    -0.16
     bil
    -0.16
    ýn
    -0.16
    POSITIVE LOGITS
     Theresa
    0.30
    Bre
    0.29
     Brexit
    0.29
     Bre
    0.27
     bre
    0.26
     Boris
    0.24
     BRE
    0.23
     Trump
    0.22
     Cummings
    0.21
     May
    0.21
    Activations Density 0.084%

    No Known Activations