© 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. 19972
    Prev
    Next
    INDEX
    Explanations

    Non-English names/words

    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
    .expect
    -0.30
    enton
    -0.28
    anel
    -0.27
    _expect
    -0.26
     mast
    -0.26
    ató
    -0.25
    ÙħÙĬر
    -0.25
    ='.
    -0.24
    errick
    -0.24
    insky
    -0.24
    POSITIVE LOGITS
    çĶĺ
    0.29
     strr
    0.28
    icit
    0.28
    æľĢåIJİä¸Ģ次
    0.27
    ç«ĸ
    0.27
    åı£æ°´
    0.26
    onis
    0.26
    å¿Ŀ
    0.26
    .inspect
    0.25
    itates
    0.25
    Activations Density 0.002%

    No Known Activations