© 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. 108722
    Prev
    Next
    INDEX
    Explanations

    scientific and technical contexts

    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
    uest
    -0.31
    é¢ĨåŁŁ
    -0.30
    hest
    -0.27
     JUST
    -0.27
     just
    -0.26
     domain
    -0.26
    éĵĿ
    -0.26
    imen
    -0.26
     Just
    -0.25
    å¼¹
    -0.25
    POSITIVE LOGITS
    åĽŀé¦ĸ
    0.30
     maté
    0.28
    Advertisements
    0.27
    erb
    0.26
    блок
    0.25
    erus
    0.25
     speeds
    0.25
    -roll
    0.25
    æ²®
    0.24
    稳
    0.24
    Activations Density 23.357%

    No Known Activations