© 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. 106943
    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
    Not in Any Lists

    No Comments

    Negative Logits
    åħļé£İ
    -0.27
    lite
    -0.25
    Lite
    -0.25
    å¹´ä¸ĭåįĬå¹´
    -0.25
     bureau
    -0.23
     Meh
    -0.23
     relatives
    -0.23
     Their
    -0.23
     dürfen
    -0.23
    ocument
    -0.23
    POSITIVE LOGITS
    ä¿¡ä»»
    0.30
    è£ħ
    0.28
    beg
    0.27
    说æĺİ书
    0.27
    æī¿
    0.26
    ä¼ij
    0.25
    Susp
    0.25
    ä¿¡
    0.25
    çİĦ
    0.25
     susp
    0.24
    Activations Density 46.074%

    No Known Activations

    This feature has no known activations.