© 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. 31-TRANSCODER-HP
    7. 17739
    Prev
    Next
    INDEX
    Explanations

    Japanese

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_31.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.31.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
     Beard
    -0.27
    Ợ
    -0.26
    é£İ湿
    -0.25
     Op
    -0.25
    娣
    -0.24
    æĸ§
    -0.24
    ç¼
    -0.24
    éĨ¯
    -0.24
    esty
    -0.24
    nosti
    -0.24
    POSITIVE LOGITS
    æĹ¥æľ¬
    0.75
    åľ¨æĹ¥æľ¬
    0.73
     Japanese
    0.72
     Japan
    0.70
     japon
    0.68
    Japanese
    0.68
     japan
    0.67
    Japan
    0.66
    æĹ¥æľ¬äºº
    0.63
     Tokyo
    0.62
    Activations Density 5.317%

    No Known Activations