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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 19-TRANSCODER-HP
    7. 4759
    Prev
    Next
    INDEX
    Explanations

    surgery

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_19.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.19.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    uur
    -0.30
    çĪ±äºº
    -0.28
     Everywhere
    -0.27
    宽æĿ¾
    -0.26
    plural
    -0.26
    umo
    -0.26
    ãĥ¼ãĤ·
    -0.25
    ibrator
    -0.25
    uras
    -0.25
    IRR
    -0.25
    POSITIVE LOGITS
     windy
    0.26
    OTH
    0.26
     ne
    0.26
     Ne
    0.25
    çĦ¶åIJİåĨį
    0.24
     clue
    0.24
    æĺ¯æ²¡æľī
    0.24
     offensive
    0.24
    ä¸įåĪ©
    0.24
    rio
    0.24
    Activations Density 0.022%

    No Known Activations