© 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. 129101
    Prev
    Next
    INDEX
    Explanations

    cause

    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
    å¾ĹçĽĬäºİ
    -0.29
     thanks
    -0.27
    ich
    -0.27
    itere
    -0.24
    å¼ı
    -0.24
    ICODE
    -0.24
    beros
    -0.24
    æĹ¥åľ¨
    -0.24
     (?)
    -0.24
     ro
    -0.24
    POSITIVE LOGITS
    ä¸ĢæĿ¯
    0.30
    è¿Ļç§įæĥħåĨµ
    0.27
    年轻人
    0.27
    éŀŃ
    0.26
    è¿ĻæĶ¯
    0.26
    è¿Ļç§įäºĭæĥħ
    0.26
    ÑĮÑı
    0.26
    à¤Ł
    0.26
    ascar
    0.25
     incompet
    0.24
    Activations Density 0.011%

    No Known Activations