© 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. 129028
    Prev
    Next
    INDEX
    Explanations

    code imports

    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
    èĩĤ
    -0.32
    çīĪ
    -0.29
    oley
    -0.27
    饱满
    -0.27
    æĹ¥çħ§
    -0.25
    rips
    -0.25
    è¿ŀæİ¥
    -0.25
    icular
    -0.25
    ä¼ı
    -0.24
    abay
    -0.24
    POSITIVE LOGITS
     period
    0.29
    idue
    0.26
    period
    0.26
     ÑĢеж
    0.25
    (phase
    0.25
    Period
    0.24
     Period
    0.24
     Cheat
    0.24
     pud
    0.24
    èĥ½å¾ĹåĪ°
    0.24
    Activations Density 0.181%

    No Known Activations