© 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. 20-TRANSCODER-HP
    7. 130930
    Prev
    Next
    INDEX
    Explanations

    Dates and numbers

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    ital
    -0.28
    ida
    -0.28
     Prior
    -0.28
    ter
    -0.27
    Prior
    -0.26
    und
    -0.26
    ung
    -0.26
    ilde
    -0.25
    ä¸Ģæĺ¯
    -0.25
     actually
    -0.25
    POSITIVE LOGITS
    .VK
    0.28
    ä¼łåĩº
    0.27
    ropri
    0.27
    åĮºåŁŁæĢ§
    0.26
    ä¾ĭåŃIJ
    0.26
    uyo
    0.25
    è´Ī
    0.25
    osit
    0.25
    enan
    0.25
    è½´
    0.25
    Activations Density 0.893%

    No Known Activations