© 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. 22-TRANSCODER-HP
    7. 130998
    Prev
    Next
    INDEX
    Explanations

    Code and technical writing excerpts

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    wick
    -0.29
    pha
    -0.28
    .ylabel
    -0.25
    ules
    -0.24
    éļIJ身
    -0.24
     secondary
    -0.24
    .locals
    -0.24
    oard
    -0.23
    ^^^^
    -0.23
    éļ±ç§ģ
    -0.23
    POSITIVE LOGITS
    åĽŀåĽ½
    0.29
    ä¿ĿçķĻ
    0.27
    天ä¸ĭ
    0.27
    deb
    0.26
     dispon
    0.26
    计åħ¥
    0.25
     retained
    0.25
    åĩ¡
    0.25
     Steel
    0.25
     ASSIGN
    0.25
    Activations Density 0.002%

    No Known Activations