© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 28956
    Prev
    Next
    INDEX
    Explanations

    references to government entities and their roles in immigration and law enforcement

    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
    maal
    -0.18
     Antworten
    -0.16
    olec
    -0.15
    UFF
    -0.15
    artial
    -0.14
     Stunden
    -0.14
     jenter
    -0.14
    arem
    -0.14
    NST
    -0.14
    .pg
    -0.14
    POSITIVE LOGITS
     Bund
    0.27
     Federal
    0.24
     federal
    0.22
    Federal
    0.21
     bund
    0.20
     land
    0.20
     Land
    0.19
    lander
    0.19
     RK
    0.18
     Ver
    0.18
    Activations Density 0.041%

    No Known Activations