© 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. 140831
    Prev
    Next
    INDEX
    Explanations

    Abbreviations and initials

    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
    ager
    -0.27
    lies
    -0.26
    sounds
    -0.26
    çĽ²
    -0.25
    н
    -0.25
    æľĢ好
    -0.25
    çİ°
    -0.25
    ivities
    -0.25
    acre
    -0.24
    åIJĮ
    -0.24
    POSITIVE LOGITS
    kbd
    0.31
    UserController
    0.26
    ï¼Įé»ĺ认
    0.26
    idUser
    0.26
    xEE
    0.25
    娼
    0.25
    ası
    0.24
    _sep
    0.24
    hammer
    0.24
    kel
    0.24
    Activations Density 3.599%

    No Known Activations