© 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. 35-TRANSCODER-HP
    7. 39428
    Prev
    Next
    INDEX
    Explanations

    mill

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_35.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.35.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ert
    -0.30
    erton
    -0.30
    idos
    -0.29
     top
    -0.28
    åı¯æĺ¯
    -0.27
    ä¸¤å¼ł
    -0.26
     range
    -0.26
    ä¹ĭ家
    -0.26
    ért
    -0.26
    ãĤ¹ãĤ¿ãĥ¼
    -0.26
    POSITIVE LOGITS
    mpp
    0.31
    nil
    0.29
    FileDialog
    0.26
    kul
    0.26
    xia
    0.26
    tableFuture
    0.25
     UIAlertView
    0.25
    Nike
    0.25
    .pp
    0.24
    æ°Ļ
    0.24
    Activations Density 0.043%

    No Known Activations