© 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. 131149
    Prev
    Next
    INDEX
    Explanations

    News articles

    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
    æĹ¢
    -0.29
     |
    -0.28
    lic
    -0.27
    da
    -0.26
    ãģĭãĤĤãģĹãĤĮãģ¾ãģĽãĤĵãģĮ
    -0.25
    ao
    -0.25
    iones
    -0.24
    auen
    -0.24
     rim
    -0.24
    æĹ¢æĺ¯
    -0.24
    POSITIVE LOGITS
     kino
    0.29
    _SB
    0.27
    RIES
    0.27
    nameof
    0.27
     partnering
    0.27
    该æĢİä¹ĪåĬŀ
    0.26
    emble
    0.25
    ,port
    0.25
    phones
    0.25
    eger
    0.25
    Activations Density 12.813%

    No Known Activations