Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 53
    Prev
    Next
    INDEX
    Explanations

    names of places or locations

    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
     Cities
    -0.20
     cities
    -0.18
    Cities
    -0.17
     City
    -0.16
     universities
    -0.16
     Universities
    -0.16
    lette
    -0.15
     city
    -0.15
    ovich
    -0.15
     Caucus
    -0.15
    POSITIVE LOGITS
     Road
    0.22
     Common
    0.20
     Rise
    0.19
     nr
    0.19
    -on
    0.18
     ROAD
    0.18
     Playing
    0.18
     Estate
    0.18
     Abbas
    0.18
    -road
    0.18
    Activations Density 0.048%

    No Known Activations