© 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. 64701
    Prev
    Next
    INDEX
    Explanations

    older

    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
     vis
    -0.30
     min
    -0.29
    om
    -0.28
    et
    -0.28
     visible
    -0.27
     man
    -0.27
    æĦı
    -0.26
     »
    -0.26
     Vis
    -0.25
     ho
    -0.25
    POSITIVE LOGITS
    纲
    0.35
    ç½ij讯
    0.28
    太å°ij
    0.26
     fucks
    0.26
    ibbon
    0.26
    oline
    0.26
    /Internal
    0.25
     african
    0.25
    è¿ĺ好
    0.25
    妲
    0.24
    Activations Density 8.734%

    No Known Activations