© 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. 33407
    Prev
    Next
    INDEX
    Explanations

    phrases related to legal accountability and intent in actions

    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
    ãģ¡ãģ¯
    -0.15
     certain
    -0.15
    obi
    -0.15
     Certain
    -0.15
     tiener
    -0.14
    omas
    -0.14
    quam
    -0.14
    rott
    -0.14
     vig
    -0.14
    aign
    -0.14
    POSITIVE LOGITS
    within
    0.32
     within
    0.32
    Within
    0.27
     Within
    0.26
    _within
    0.22
     covered
    0.22
     Covered
    0.20
     entitled
    0.18
     pretext
    0.17
    -covered
    0.17
    Activations Density 0.080%

    No Known Activations