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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 55197
    Prev
    Next
    INDEX
    Explanations

    references to political figures and their actions in legislative contexts

    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
    Democratic
    -0.21
     Democrat
    -0.21
     Democratic
    -0.21
     Democrats
    -0.19
     Liberals
    -0.19
     bipartisan
    -0.18
     democratic
    -0.18
     Liberal
    -0.17
     Democr
    -0.17
     democr
    -0.17
    POSITIVE LOGITS
     Tea
    0.39
     tea
    0.35
    tea
    0.30
    Te
    0.24
     purity
    0.23
     teas
    0.23
     TE
    0.22
    èĮ¶
    0.21
     te
    0.20
     Conserv
    0.19
    Activations Density 0.105%

    No Known Activations