© 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. 35-TRANSCODER-HP
    7. 39204
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     sire
    -0.25
    ''.
    -0.24
     fully
    -0.24
    ä¸ĢæĿ¡
    -0.24
    '',
    -0.24
    ä¸Ģç¢Ĺ
    -0.23
    æ¿ĢåĬ±
    -0.23
    dw
    -0.23
    "
    -0.23
    otti
    -0.22
    POSITIVE LOGITS
    ä¹Łä¸įä¾ĭå¤ĸ
    0.30
    饼
    0.29
    pery
    0.29
    MOST
    0.27
    èĥĸ
    0.25
    éĿ¡
    0.25
    icals
    0.24
    æİĢ
    0.24
    setBackground
    0.24
    æijĺç¼ĸ
    0.23
    Activations Density 3.241%

    No Known Activations

    This feature has no known activations.