© 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. 108532
    Prev
    Next
    INDEX
    Explanations

    below

    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
     Seasons
    -0.30
    semb
    -0.27
    åħļçļĦ建设
    -0.26
    åľ¨åħ¶
    -0.26
    usher
    -0.25
    جاد
    -0.24
    æĸ°æĿIJæĸĻ
    -0.24
    æĸ°çļĦ
    -0.24
    ewn
    -0.24
    çĪ±ç¾İ
    -0.24
    POSITIVE LOGITS
    ç¼ĸ
    0.26
    ç²Ĺ
    0.25
    è¿ľ
    0.25
    ovich
    0.25
    æĢł
    0.25
     register
    0.24
     registers
    0.24
    Api
    0.24
    profiles
    0.23
     tele
    0.23
    Activations Density 0.565%

    No Known Activations