© 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. 20-TRANSCODER-HP
    7. 108650
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    Negative Logits
    åĮį
    -0.28
    èĢĮæĪIJ
    -0.26
    æ»ļæ»ļ
    -0.26
    Boost
    -0.26
    FOUNDATION
    -0.25
    eler
    -0.25
     ká»ĭp
    -0.24
     quilt
    -0.23
    éĢļè¯Ŀ
    -0.23
    .persistent
    -0.23
    POSITIVE LOGITS
     tower
    0.29
    /design
    0.26
     easily
    0.25
     advance
    0.25
     advanced
    0.25
     imag
    0.24
    æĺĵäºİ
    0.24
    é«ĺ级
    0.24
    .linear
    0.24
    SPORT
    0.24
    Activations Density 8.687%

    No Known Activations

    This feature has no known activations.