© 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. 108555
    Prev
    Next
    INDEX
    Explanations

    code

    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
    ä¹Łä¸įæķ¢
    -0.26
    ton
    -0.25
     afs
    -0.25
    ogue
    -0.24
    æĮĩ示
    -0.24
    sist
    -0.24
    лив
    -0.24
    åijľ
    -0.24
     CONTRIBUTORS
    -0.24
    arker
    -0.24
    POSITIVE LOGITS
    ijd
    0.29
    DDD
    0.26
    è¾ĩ
    0.26
    åħ«åį¦
    0.26
    å¤ļè¾¾
    0.25
    dig
    0.25
    æĩ
    0.25
    ÑģоÑĢ
    0.25
    edException
    0.25
     Bonus
    0.24
    Activations Density 12.209%

    No Known Activations