© 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. 41642
    Prev
    Next
    INDEX
    Explanations

    occurrences of the phrase "food and drink" along with related regulatory terms

    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
     Foods
    -0.16
     foods
    -0.16
     destructor
    -0.15
    èĢ³
    -0.15
    ator
    -0.15
    ç¸
    -0.14
    ôt
    -0.14
    ights
    -0.14
    prus
    -0.14
    aro
    -0.14
    POSITIVE LOGITS
     drink
    0.34
     beverage
    0.30
     Drink
    0.29
     Beverage
    0.29
     Bever
    0.27
     beverages
    0.25
     drinks
    0.25
     bev
    0.24
    Drink
    0.23
     Drinks
    0.23
    Activations Density 0.008%

    No Known Activations