© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 20-TRANSCODER-HP
    7. 140981
    Prev
    Next
    INDEX
    Explanations

    [

    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
    Not in Any Lists

    No Comments

    Negative Logits
    综åIJĪ
    -0.31
    man
    -0.28
     challenging
    -0.26
    é¦
    -0.25
     joined
    -0.24
    VIP
    -0.24
    綾åIJĪ
    -0.24
    综åIJĪæĢ§
    -0.24
     intent
    -0.24
     attempts
    -0.24
    POSITIVE LOGITS
    ctions
    0.28
     kontakt
    0.27
    ä¸įç»ı
    0.26
    ä»»ä½ķä¸Ģ个
    0.26
    å·²ç»ıè¾¾åĪ°
    0.25
    ÑĨен
    0.25
    ÑįÑĤ
    0.24
    åģľæĶ¾
    0.24
    äch
    0.24
     nudity
    0.24
    Activations Density 7.921%

    No Known Activations