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. 23660
    Prev
    Next
    INDEX
    Explanations

    references to privilege and its impacts

    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
    tract
    -0.17
    rey
    -0.16
     Zucker
    -0.15
    ropolis
    -0.15
    stras
    -0.14
    rec
    -0.14
    OTA
    -0.14
    ow
    -0.14
    elf
    -0.13
    iverse
    -0.13
    POSITIVE LOGITS
    ously
    0.23
    ilege
    0.21
     privilege
    0.21
     Priv
    0.21
     priv
    0.20
    ileges
    0.19
    (priv
    0.18
    iferay
    0.18
    priv
    0.18
    ately
    0.17
    Activations Density 0.007%

    No Known Activations