© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 30-TRANSCODER-HP
    7. 95076
    Prev
    Next
    INDEX
    Explanations

    hyphen

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_30.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.30.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    们
    -0.47
    们çļĦ
    -0.43
    们éĥ½
    -0.39
    åĢij
    -0.37
    ëĵ¤ìĿĺ
    -0.31
    ows
    -0.30
    rients
    -0.30
    sWith
    -0.30
    ues
    -0.29
    ibles
    -0.28
    POSITIVE LOGITS
    è·ĥ
    0.29
     Medicine
    0.26
    Music
    0.24
     Coupe
    0.24
    Swagger
    0.23
    ppe
    0.23
    ä¸ĢèĦ¸
    0.23
    (TM
    0.23
    vre
    0.23
    åħŃåIJĪ
    0.23
    Activations Density 0.011%

    No Known Activations