© 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. 28844
    Prev
    Next
    INDEX
    Explanations

    keywords related to protests, substance use, and team dynamics

    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
     tomorrow
    -0.23
     today
    -0.21
     afterward
    -0.21
     aujourd
    -0.20
     heute
    -0.20
     hoy
    -0.19
    iland
    -0.18
     afterwards
    -0.18
    oday
    -0.18
    ä»Ĭ天
    -0.18
    POSITIVE LOGITS
     earlier
    0.79
     Earlier
    0.61
     previous
    0.60
    Earlier
    0.58
     previously
    0.53
     Previous
    0.46
    Previous
    0.46
    previous
    0.46
     prior
    0.43
     Previously
    0.41
    Activations Density 0.054%

    No Known Activations