© 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. 142619
    Prev
    Next
    INDEX
    Explanations

    Code and repeated 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
    -know
    -0.27
    åľ¨å¤ĸ
    -0.26
     NAN
    -0.25
    å¾Ĺ天
    -0.25
    spy
    -0.24
    åįģå¹´
    -0.24
     independently
    -0.24
    ulative
    -0.24
    ר×Ĵ
    -0.23
    å®Įåħ¨
    -0.23
    POSITIVE LOGITS
    nex
    0.32
    åĨĮ
    0.26
     dash
    0.25
    ix
    0.24
     Citizenship
    0.24
     Dix
    0.24
    French
    0.23
     [~,
    0.23
     myself
    0.23
    UX
    0.23
    Activations Density 1.875%

    No Known Activations