© 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. 33152
    Prev
    Next
    INDEX
    Explanations

    references to George W. Bush

    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
    inidad
    -0.15
    orage
    -0.14
    ocate
    -0.14
    ovÃŃ
    -0.13
    ÃĸL
    -0.13
    %C
    -0.13
     wake
    -0.13
    QUIRE
    -0.13
    aye
    -0.12
    endo
    -0.12
    POSITIVE LOGITS
     Bush
    0.43
    Bush
    0.36
     bush
    0.32
     bushes
    0.18
     Dub
    0.16
    administration
    0.15
    Dub
    0.15
    -Trump
    0.15
     Reagan
    0.15
     Cheney
    0.14
    Activations Density 0.010%

    No Known Activations