© 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. 22-TRANSCODER-HP
    7. 98297
    Prev
    Next
    INDEX
    Explanations

    em dashes

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ä¸ĵä¸ļçļĦ
    -0.26
    Towards
    -0.26
    (ai
    -0.25
    sett
    -0.25
    köp
    -0.24
     known
    -0.24
    ảo
    -0.24
    quam
    -0.24
     key
    -0.23
    obel
    -0.23
    POSITIVE LOGITS
     unsuccessfully
    0.30
    istribute
    0.27
    è·ijäºĨ
    0.27
     unnecessarily
    0.27
    ä¸ĢéĿ¢
    0.27
    è¾Ł
    0.26
    æĢ§åĴĮ
    0.26
    ä¸Ģç¯ĩæĸĩç«ł
    0.26
    å¹´ä¸ŃåĽ½
    0.25
    -initialized
    0.25
    Activations Density 0.008%

    No Known Activations