© 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. 13228
    Prev
    Next
    INDEX
    Explanations

    mentions of food-related events and experiences

    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
    ActionCreators
    -0.16
     regular
    -0.14
    trand
    -0.14
    ritz
    -0.14
    cona
    -0.14
     Rosenstein
    -0.14
    typeid
    -0.14
    /stdc
    -0.13
     demand
    -0.13
     combust
    -0.13
    POSITIVE LOGITS
     sampling
    0.40
     tasting
    0.39
     samples
    0.35
     Sampling
    0.33
     sample
    0.31
    Sampling
    0.28
     sampler
    0.28
     sampled
    0.28
    .sample
    0.28
     Samples
    0.27
    Activations Density 0.045%

    No Known Activations