© 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. 63831
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    è¿Ļä¹Ī大
    -0.26
    nid
    -0.24
    ød
    -0.24
     Dunn
    -0.24
    åĢª
    -0.24
    idan
    -0.23
     zij
    -0.23
    æĺ¯ä»ĸ
    -0.23
     evidently
    -0.23
    对人ä½ĵ
    -0.23
    POSITIVE LOGITS
    æī¿
    0.29
     Colt
    0.26
    cient
    0.25
    trad
    0.25
    thal
    0.24
    æį¡
    0.24
    utors
    0.24
     respect
    0.24
    æijĶ
    0.23
    cobra
    0.23
    Activations Density 4.936%

    No Known Activations

    This feature has no known activations.