© 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. 130387
    Prev
    Next
    INDEX
    Explanations

    most

    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
    ç¥Ī
    -0.28
    娴
    -0.25
    odate
    -0.25
    emie
    -0.25
    es
    -0.24
    ä¿Ŀè´¹
    -0.24
    atee
    -0.24
    inan
    -0.24
    åĽłåľ°
    -0.24
    rieg
    -0.23
    POSITIVE LOGITS
    åĩłä¹İ
    0.30
    å¹¾ä¹İ
    0.29
    åĩłä¹İæĺ¯
    0.27
    ç»Ŀ大å¤ļæķ°
    0.27
    è½®
    0.27
    åĩłä¹İæīĢæľī
    0.27
     majority
    0.26
     hầu
    0.26
     almost
    0.26
     Nearly
    0.26
    Activations Density 8.407%

    No Known Activations