© 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. 108651
    Prev
    Next
    INDEX
    Explanations

    evaluation, reports, studies

    np_max-act · gemini-2.0-flash
    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
    omanip
    -0.29
    INI
    -0.26
    oley
    -0.25
    大éĻĨ
    -0.25
     Patton
    -0.25
    Substring
    -0.24
    ération
    -0.23
     Ih
    -0.23
     INCLUDED
    -0.23
    DUCT
    -0.23
    POSITIVE LOGITS
     Wales
    0.30
    è¯Ŀ
    0.28
     Lect
    0.27
    beta
    0.25
    ÙħاÙħ
    0.25
    ç«Ļ
    0.24
    claimer
    0.24
     Debate
    0.24
    éĩİ
    0.23
    ves
    0.23
    Activations Density 14.252%

    No Known Activations