© 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. 32-TRANSCODER-HP
    7. 40950
    Prev
    Next
    INDEX
    Explanations

    Number five

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    riba
    -0.32
    ños
    -0.30
    ç¾ģ
    -0.25
    ä¸įèµ·
    -0.25
    chè
    -0.24
     Third
    -0.24
    å¾®
    -0.24
    ä¸ĩæĪ·
    -0.24
    并ä¸į代表
    -0.24
    æģ¯æģ¯
    -0.24
    POSITIVE LOGITS
     pent
    0.30
    pent
    0.30
    änder
    0.29
    -five
    0.28
    äºĶ个
    0.27
    5
    0.27
    OND
    0.26
    five
    0.26
    cono
    0.26
     five
    0.26
    Activations Density 0.012%

    No Known Activations