© 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. 128957
    Prev
    Next
    INDEX
    Explanations

    Code/document text fragments

    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
    изм
    -0.29
    åĵģ
    -0.28
    ç»Ĩç»Ĩ
    -0.27
    åĬŀäºĭ
    -0.26
    s
    -0.25
    éĩį
    -0.25
    atos
    -0.25
    ä½ıæĪ¿åĴĮ
    -0.24
    æľ«
    -0.24
    磼
    -0.24
    POSITIVE LOGITS
    è¦ģçŁ¥éģĵ
    0.32
    åĽłä¸ºä»ĸ
    0.27
    orta
    0.27
    ç»Ļä»ĸ
    0.26
     swallow
    0.25
    åĽłä¸ºä½ł
    0.25
    OUTH
    0.25
    _DIRECT
    0.24
    errat
    0.24
    èѬå¦Ĥ
    0.24
    Activations Density 0.028%

    No Known Activations