© 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. 63482
    Prev
    Next
    INDEX
    Explanations

    phrases related to law enforcement actions and citations

    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
    ocal
    -0.17
    ause
    -0.14
    æ·»
    -0.14
     pož
    -0.14
     dood
    -0.13
     touched
    -0.13
    ãĥ¡ãĥ³ãĥĪ
    -0.13
    ">//
    -0.13
    æŁĵ
    -0.13
    åĽ²
    -0.13
    POSITIVE LOGITS
     ticket
    0.54
     tickets
    0.50
     Tickets
    0.45
     Ticket
    0.44
     citation
    0.42
    -ticket
    0.41
     citations
    0.40
    ticket
    0.40
    tickets
    0.39
    Ticket
    0.38
    Activations Density 0.045%

    No Known Activations