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. 31347
    Prev
    Next
    INDEX
    Explanations

    instances of public statements or remarks made by political figures

    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
    anson
    -0.15
    /sources
    -0.15
    ouser
    -0.14
    alue
    -0.14
    ²
    -0.14
    maries
    -0.13
    annes
    -0.13
    QualifiedName
    -0.13
    лаж
    -0.13
    ALAR
    -0.13
    POSITIVE LOGITS
     words
    0.84
    words
    0.66
     Words
    0.60
    Words
    0.56
    _words
    0.50
     WORD
    0.48
     palabras
    0.48
     word
    0.46
    WORDS
    0.45
    .words
    0.45
    Activations Density 0.042%

    No Known Activations