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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 24806
    Prev
    Next
    INDEX
    Explanations

    references to cultural identity and shared heritage

    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
     realms
    -0.15
    cmp
    -0.13
    aggio
    -0.13
    åĬĽéĩı
    -0.13
    λÏħ
    -0.13
    956
    -0.13
    nesty
    -0.13
    ниÑģÑĤ
    -0.13
     бл
    -0.13
     hạng
    -0.12
    POSITIVE LOGITS
     culture
    0.75
     Culture
    0.58
    culture
    0.58
    Culture
    0.52
     cultures
    0.43
    æĸĩåĮĸ
    0.41
     customs
    0.41
     cultura
    0.40
    ulture
    0.37
     cultural
    0.37
    Activations Density 0.087%

    No Known Activations