© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 41347
    Prev
    Next
    INDEX
    Explanations

    discussions surrounding societal inequities and demographic characteristics

    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
    onas
    -0.15
    ,
    -0.14
    ona
    -0.14
    нод
    -0.14
    -0.14
    _
    -0.14
    esch
    -0.13
    ibase
    -0.13
    ombok
    -0.13
     occurring
    -0.13
    POSITIVE LOGITS
     comes
    0.53
     come
    0.53
    come
    0.45
     Come
    0.44
    comes
    0.43
     Comes
    0.43
    ä¾Ĩ
    0.43
    Come
    0.43
     came
    0.41
    æĿ¥èĩª
    0.40
    Activations Density 0.053%

    No Known Activations