© 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. Qwen3-4B
    5. 20-TRANSCODER-HP
    7. 129746
    Prev
    Next
    INDEX
    Explanations

    -based

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    âĢ¦)
    -0.29
    itudes
    -0.29
    tures
    -0.28
     boosting
    -0.26
    ulence
    -0.26
    orough
    -0.25
    ä¸įåIJĮæĦı
    -0.25
     _$
    -0.25
    ENCY
    -0.24
    å°Ķæĸ¯
    -0.24
    POSITIVE LOGITS
    çĦ¶åIJİåĨį
    0.27
     graphs
    0.26
    rix
    0.25
    AXB
    0.24
    èĢĮåĩº
    0.23
    anter
    0.23
    -dependent
    0.23
    çī©ä»·
    0.23
    -based
    0.23
     BAM
    0.23
    Activations Density 6.088%

    No Known Activations