© 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. 88950
    Prev
    Next
    INDEX
    Explanations

    Taking control/capture

    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
    conomy
    -0.28
    æ´Ĺ澡
    -0.27
    æ³Ľ
    -0.27
    æĺ¥å¤©
    -0.25
    "..
    -0.25
    æĪĸ许æĺ¯
    -0.25
     recognition
    -0.24
    classnames
    -0.24
    _prom
    -0.23
     paradox
    -0.23
    POSITIVE LOGITS
     пол
    0.30
    ä¿ĺ
    0.30
    æĬ¢
    0.26
    Åĺ
    0.26
    æIJ¶
    0.26
    æĵĴ
    0.26
    æİ¥
    0.26
     nab
    0.25
     Pol
    0.25
    ewhat
    0.25
    Activations Density 0.008%

    No Known Activations