© 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. 19-TRANSCODER-HP
    7. 9661
    Prev
    Next
    INDEX
    Explanations

    General text

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    Dirty
    -0.28
    IMA
    -0.26
    Builder
    -0.25
     üye
    -0.25
     Mystery
    -0.24
    ÇĴ
    -0.24
     tÆ°
    -0.24
     Favorites
    -0.24
     trium
    -0.24
    æ¶Įçİ°
    -0.24
    POSITIVE LOGITS
    upy
    0.30
    isha
    0.29
    fts
    0.28
    èĢĥåľº
    0.28
     gag
    0.26
    ESH
    0.26
     ung
    0.26
    ocr
    0.26
    ést
    0.25
     esc
    0.24
    Activations Density 0.019%

    No Known Activations