© 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. 19918
    Prev
    Next
    INDEX
    Explanations

    span

    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
     cov
    -0.28
    oval
    -0.28
    جÙĦ
    -0.27
    éĢīåĩº
    -0.25
     BrowserModule
    -0.25
     autonomous
    -0.24
    Aut
    -0.24
     comb
    -0.24
     Eclipse
    -0.24
     AppRoutingModule
    -0.24
    POSITIVE LOGITS
     span
    0.53
     spans
    0.48
    Span
    0.47
     Span
    0.46
    span
    0.42
    _span
    0.41
     spanning
    0.40
    跨度
    0.38
    .span
    0.37
    (span
    0.35
    Activations Density 4.888%

    No Known Activations