© 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. 17858
    Prev
    Next
    INDEX
    Explanations

    Legal/Financial/Administrative texts

    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
    èµĦæľ¬
    -0.33
     capital
    -0.31
     капиÑĤ
    -0.31
    帽
    -0.31
     Capital
    -0.31
     capitals
    -0.30
    èµĦéĩij
    -0.30
     capitalize
    -0.30
    money
    -0.29
    capital
    -0.28
    POSITIVE LOGITS
     augment
    0.29
     Aug
    0.29
    èĢĮå·²
    0.28
    ste
    0.27
    深度
    0.27
    [Unit
    0.27
    aug
    0.26
    ofile
    0.26
     Jul
    0.26
     Rolls
    0.26
    Activations Density 0.246%

    No Known Activations