© 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. 64829
    Prev
    Next
    INDEX
    Explanations

    punctuation

    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
    addle
    -0.32
    IGATION
    -0.27
    adena
    -0.27
    leur
    -0.27
    ä¸ŃåĽ½åĪ¶éĢł
    -0.26
    åı¯è¾¾
    -0.26
    ilha
    -0.26
    è¿©
    -0.25
    ologue
    -0.24
    ëĵĿ
    -0.24
    POSITIVE LOGITS
     signific
    0.26
    ãģĻãģĶãģĦ
    0.26
    å¥ĩ
    0.26
    åIJĦ级
    0.26
    羡
    0.25
     extraordinary
    0.25
    羣æĺ¯
    0.25
    ãģĬãģĭ
    0.25
    éĿŀ常çļĦ
    0.24
    LY
    0.24
    Activations Density 15.418%

    No Known Activations