© 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. 25-TRANSCODER-HP
    7. 108680
    Prev
    Next
    INDEX
    Explanations

    foreign character strings

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_25.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.25.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     beta
    -0.26
    ndx
    -0.25
     SHOP
    -0.25
     overlaps
    -0.25
     held
    -0.25
    Next
    -0.24
     overlap
    -0.24
    éħIJ
    -0.24
    bucks
    -0.24
     Next
    -0.24
    POSITIVE LOGITS
    åį¡çīĩ
    0.31
     BaseController
    0.30
    .lu
    0.28
    æĥ©ç½ļ
    0.28
    æī¹ç¤º
    0.27
    .sent
    0.27
    åĬ³åĬ¨
    0.26
    åĩºåħ¥å¢ĥ
    0.26
    æĻ®éĢļ人
    0.25
    åĪĨéĴŁåIJİ
    0.24
    Activations Density 0.946%

    No Known Activations