© 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. 22220
    Prev
    Next
    INDEX
    Explanations

    terms related to crime and types of numerical reasoning

    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
     nÃło
    -0.16
    |array
    -0.16
    tings
    -0.16
    ÅĻ
    -0.15
    erson
    -0.15
    odore
    -0.15
    TING
    -0.14
    ÑĤин
    -0.14
    à¹Ģหล
    -0.14
    esc
    -0.14
    POSITIVE LOGITS
    .ArgumentParser
    0.21
    ipel
    0.20
    ansas
    0.19
    enal
    0.19
    bitrary
    0.19
    quivos
    0.19
    adia
    0.17
    itecture
    0.17
    åĬ¿
    0.17
    itect
    0.17
    Activations Density 0.043%

    No Known Activations