© 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. 111993
    Prev
    Next
    INDEX
    Explanations

    become

    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
     rank
    -0.25
     form
    -0.25
    æİī
    -0.24
     onto
    -0.24
    hol
    -0.24
     ÑįÑĤомÑĥ
    -0.24
    åįķ
    -0.24
    .GetAsync
    -0.24
    ä¹İ
    -0.24
    èĶ¼
    -0.23
    POSITIVE LOGITS
     synonymous
    0.31
    ä¸Ģ家人
    0.29
    ä¸ĢåĽ¢
    0.29
     fodder
    0.28
    pire
    0.26
    çļĦä¸Ģåijĺ
    0.26
    è¦ģ说
    0.26
     part
    0.26
    åij¨åĽĽ
    0.25
    æĢİä¹ĪåĽŀäºĭ
    0.25
    Activations Density 0.054%

    No Known Activations