© 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. 142632
    Prev
    Next
    INDEX
    Explanations

    punctuation

    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
    andes
    -0.31
    armed
    -0.30
    zap
    -0.28
     pyt
    -0.28
     tapes
    -0.26
     prince
    -0.26
    Cop
    -0.26
    bett
    -0.25
     taped
    -0.24
    kop
    -0.24
    POSITIVE LOGITS
    gnore
    0.30
    ellow
    0.29
    æĹłå¿§
    0.27
    olver
    0.26
    æŁ¥çľĭæĽ´å¤ļ
    0.25
    ä¸Ģè§Ī
    0.25
     Mang
    0.24
    oplast
    0.24
    clair
    0.24
    ñe
    0.24
    Activations Density 0.913%

    No Known Activations