© 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. 63597
    Prev
    Next
    INDEX
    Explanations

    categories and groupings related to arrests and offenses

    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
    ahlen
    -0.16
    uzzi
    -0.14
    ellation
    -0.14
     kok
    -0.14
     experiment
    -0.14
    uco
    -0.14
    decess
    -0.14
    arez
    -0.13
    iple
    -0.13
    mav
    -0.13
    POSITIVE LOGITS
     category
    0.74
     categories
    0.67
     Category
    0.64
    category
    0.61
    Category
    0.59
     Categories
    0.57
    .category
    0.55
    _category
    0.54
    -category
    0.54
    categories
    0.53
    Activations Density 0.033%

    No Known Activations