© 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. 23815
    Prev
    Next
    INDEX
    Explanations

    references to Hollywood and star-related contexts, particularly related to Donald Trump and his star on the Hollywood Walk of Fame

    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
    _ABI
    -0.17
     InitializeComponent
    -0.15
    åı¤
    -0.15
    .opensource
    -0.14
     Straw
    -0.14
    etÃŃ
    -0.14
    稿
    -0.14
    ĶåĽŀ
    -0.14
    abbit
    -0.14
    ikon
    -0.14
    POSITIVE LOGITS
     plaque
    0.34
     pla
    0.30
     marker
    0.26
     bronze
    0.26
     markers
    0.25
     inscription
    0.24
     dedic
    0.23
     granite
    0.23
     dedication
    0.22
    à¹ģà¸ľ
    0.22
    Activations Density 0.023%

    No Known Activations