© 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. 64802
    Prev
    Next
    INDEX
    Explanations

    house

    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
    ties
    -0.30
    大家åĪĨ享
    -0.27
    è¿IJæ°Ķ
    -0.26
    tır
    -0.25
     Raphael
    -0.24
    xDF
    -0.24
     Dum
    -0.24
     tức
    -0.23
     Harmon
    -0.23
     Dodd
    -0.23
    POSITIVE LOGITS
     diss
    0.26
    çŃij
    0.25
    æIJŃ
    0.24
    é³³
    0.23
     prer
    0.23
    è¦ģçŁ¥éģĵ
    0.23
     trait
    0.23
    ined
    0.23
    åī¯
    0.23
     Mae
    0.23
    Activations Density 12.567%

    No Known Activations