© 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. 62713
    Prev
    Next
    INDEX
    Explanations

    Numbers and dates

    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
    ä»İä¸ļ人åijĺ
    -0.28
    åĴ«
    -0.26
     Parm
    -0.26
    enga
    -0.26
    åIJĮçŃī
    -0.25
    umbs
    -0.25
    '&&
    -0.25
    åIJĮæĹ¶ä¹Łæĺ¯
    -0.25
     Nagar
    -0.25
    ç½ij讯
    -0.25
    POSITIVE LOGITS
    orea
    0.29
    unkt
    0.26
    壮
    0.25
    imitive
    0.24
    rollo
    0.24
    .bp
    0.24
    æ²»
    0.24
     mounts
    0.23
    åĨį
    0.23
     notices
    0.23
    Activations Density 6.788%

    No Known Activations