© 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. 33237
    Prev
    Next
    INDEX
    Explanations

    references to quantitative statistics or estimations regarding individuals, particularly women

    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
    rada
    -0.15
    etine
    -0.13
    odzi
    -0.13
     hierarchy
    -0.12
     ê¿Ī
    -0.12
    TestFixture
    -0.12
    ½Ķ
    -0.12
    lobber
    -0.12
    chte
    -0.12
    _predicted
    -0.12
    POSITIVE LOGITS
     count
    1.11
     Count
    1.00
     counting
    1.00
     counted
    0.96
    count
    0.96
     counts
    0.95
    -count
    0.95
     COUNT
    0.92
    Count
    0.87
    .count
    0.85
    Activations Density 0.014%

    No Known Activations