© 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. 62945
    Prev
    Next
    INDEX
    Explanations

    Global news and tech connections

    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
    Not in Any Lists

    No Comments

    Negative Logits
    .dex
    -0.28
     itself
    -0.27
     PPC
    -0.25
    稳
    -0.25
    丼
    -0.24
    -package
    -0.24
    对åħ¶çľŁå®ŀ
    -0.23
    iture
    -0.23
    åħļæĶ¿
    -0.23
    packed
    -0.23
    POSITIVE LOGITS
    prot
    0.28
    èµ°
    0.28
    Ø¡
    0.28
    åĵģ
    0.27
    -forward
    0.27
    edd
    0.26
    ummer
    0.26
     пÑĢоÑĤ
    0.26
    æĭĸ
    0.26
    Specifications
    0.26
    Activations Density 13.932%

    No Known Activations