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. 23302
    Prev
    Next
    INDEX
    Explanations

    references to environmental concerns, particularly related to water contamination and pollution

    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
     double
    -0.24
     Double
    -0.23
    -double
    -0.22
    Double
    -0.20
    double
    -0.19
    .nextDouble
    -0.18
     DOUBLE
    -0.18
     doubled
    -0.16
    VRT
    -0.16
    ,double
    -0.16
    POSITIVE LOGITS
     floating
    0.37
     float
    0.36
     buoy
    0.36
     floats
    0.35
     Floating
    0.31
     raft
    0.31
     Float
    0.31
     bob
    0.30
     flotation
    0.30
     FLOAT
    0.30
    Activations Density 0.011%

    No Known Activations