© 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. 19917
    Prev
    Next
    INDEX
    Explanations

    code/documentation

    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
    itelist
    -0.28
     rico
    -0.28
     беÑģплаÑĤно
    -0.27
     politely
    -0.25
    arc
    -0.25
    åĨįå¤ļ
    -0.25
    æģ©
    -0.24
    ChartData
    -0.24
    LOPT
    -0.24
     rdr
    -0.24
    POSITIVE LOGITS
    inan
    0.26
    fts
    0.25
    ystone
    0.25
    [end
    0.25
    endo
    0.25
    æ¹ĸ
    0.24
     burn
    0.24
    uding
    0.23
     refinement
    0.23
    <Entry
    0.23
    Activations Density 1.164%

    No Known Activations