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. 23504
    Prev
    Next
    INDEX
    Explanations

    references to deception or imitation in a context involving music or art

    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
    linger
    -0.17
    ushman
    -0.16
    Unchecked
    -0.15
    TexParameter
    -0.15
    ãĥ¼ãĥį
    -0.14
     ká»
    -0.14
    achts
    -0.13
    /Foundation
    -0.13
    icho
    -0.13
    á»ijc
    -0.13
    POSITIVE LOGITS
     fake
    0.69
     Fake
    0.57
    fake
    0.56
    Fake
    0.54
     false
    0.50
     faker
    0.47
    false
    0.41
     fals
    0.41
    (fake
    0.40
     False
    0.39
    Activations Density 0.034%

    No Known Activations