© 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. 13572
    Prev
    Next
    INDEX
    Explanations

    legal terms related to justice and civil liberties

    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
     Omaha
    -0.21
     Nebraska
    -0.19
     Illinois
    -0.17
     Connecticut
    -0.17
     Ohio
    -0.16
     Fargo
    -0.16
    azed
    -0.16
     Pens
    -0.15
    ãģIJ
    -0.15
    .US
    -0.15
    POSITIVE LOGITS
     Vancouver
    0.62
     BC
    0.51
    ancouver
    0.47
     Canucks
    0.45
    BC
    0.41
    .bc
    0.40
     Surrey
    0.39
     British
    0.38
     Fraser
    0.36
     RCMP
    0.36
    Activations Density 0.060%

    No Known Activations