© 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. 142546
    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
     airspace
    -0.27
     Publication
    -0.24
    é¢IJ
    -0.24
     BaseEntity
    -0.24
    ieu
    -0.24
    (dto
    -0.23
    RH
    -0.23
    atus
    -0.23
    .tiles
    -0.23
     Masks
    -0.23
    POSITIVE LOGITS
    DownList
    0.28
    otropic
    0.28
     affirm
    0.26
    GOP
    0.26
    ething
    0.25
     finishing
    0.24
    orary
    0.24
    ALAR
    0.24
    igned
    0.24
    usal
    0.24
    Activations Density 0.064%

    No Known Activations

    This feature has no known activations.