© 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. 130943
    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
    æłĬ
    -0.28
     <![
    -0.27
    loquent
    -0.27
     Pols
    -0.26
    anging
    -0.25
     Php
    -0.24
     Manor
    -0.24
     Tomb
    -0.24
    strt
    -0.24
    èѬå¦Ĥ
    -0.24
    POSITIVE LOGITS
    çºł
    0.29
    æĿ¨
    0.28
    å¼ł
    0.27
    èµµ
    0.27
    anner
    0.26
    éĴŁ
    0.26
     Zones
    0.26
     Ze
    0.25
    tem
    0.25
     horns
    0.25
    Activations Density 12.527%

    No Known Activations