© 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. 64610
    Prev
    Next
    INDEX
    Explanations

    Unstructured text snippets

    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
    cient
    -0.28
    cı
    -0.27
    finity
    -0.25
    Worker
    -0.25
    令
    -0.24
    cli
    -0.24
    orrent
    -0.23
    éĵ¿
    -0.23
     Worker
    -0.23
     diy
    -0.22
    POSITIVE LOGITS
    .cljs
    0.28
    /dc
    0.27
     gord
    0.26
    TRL
    0.25
    çĿĢçľ¼
    0.25
    _ble
    0.25
     gap
    0.25
    ä¸ĩä½Ļ
    0.24
    stery
    0.24
    agne
    0.24
    Activations Density 0.285%

    No Known Activations