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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 15704
    Prev
    Next
    INDEX
    Explanations

    references to bans or restrictions, particularly in a legal or policy context

    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
    indices
    -0.14
     ÙħÙĤدÙħ
    -0.14
    oot
    -0.14
     Shed
    -0.13
     Appointment
    -0.13
    agas
    -0.13
     Indices
    -0.13
    rin
    -0.12
     indices
    -0.12
    eron
    -0.12
    POSITIVE LOGITS
     ban
    0.54
    ban
    0.41
     policy
    0.38
     rule
    0.37
     decision
    0.35
     Ban
    0.34
     restriction
    0.33
     bans
    0.32
     measure
    0.30
    Ban
    0.30
    Activations Density 0.146%

    No Known Activations