© 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. 129264
    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
    å¤ļå¤ļ
    -0.27
    xFFFFFFFF
    -0.27
    leo
    -0.24
    xdd
    -0.24
     Fet
    -0.24
    ieties
    -0.23
    _Char
    -0.23
    icum
    -0.23
     Subjects
    -0.23
    Om
    -0.23
    POSITIVE LOGITS
    é¾Ħ
    0.30
    DataTask
    0.28
     liber
    0.28
    _library
    0.27
    _hash
    0.27
    éĺ¡
    0.26
    齡
    0.26
    ockets
    0.25
    olk
    0.25
    -header
    0.24
    Activations Density 2.238%

    No Known Activations

    This feature has no known activations.