© 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. 19833
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
     Integral
    -0.29
     Newcastle
    -0.27
    inspace
    -0.25
     Rift
    -0.25
    çļİ
    -0.25
    转æĬĺ
    -0.24
    azine
    -0.24
    utoff
    -0.23
    uye
    -0.23
    .osgi
    -0.23
    POSITIVE LOGITS
    楼梯
    0.28
     MÃł
    0.25
    want
    0.24
    çİĭåŃIJ
    0.24
     passwd
    0.24
    è°Ķ
    0.24
    çĶŁæ´»è´¨éĩı
    0.24
    åŁºäºİ
    0.24
    clo
    0.23
    pais
    0.23
    Activations Density 0.002%

    No Known Activations

    This feature has no known activations.