© 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. 129201
    Prev
    Next
    INDEX
    Explanations

    Article excerpts/quotations

    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
     Barcode
    -0.28
     Leak
    -0.27
     checkpoints
    -0.26
    ukes
    -0.25
     Dice
    -0.24
    _scal
    -0.24
    éĻĽ
    -0.24
    arah
    -0.24
    ãĥijãĥ¼ãĥĨãĤ£ãĥ¼
    -0.24
    (chars
    -0.24
    POSITIVE LOGITS
    æ¤Ł
    0.26
    绽
    0.25
    onian
    0.25
    æķ£åıij
    0.25
     duy
    0.24
    å¸ħ
    0.24
    åı³ä¾§
    0.23
    ä¸Ģè§Ĵ
    0.23
    åľºåĿĩ
    0.23
    eyond
    0.23
    Activations Density 18.773%

    No Known Activations