© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 22-TRANSCODER-HP
    7. 111010
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    Negative Logits
    çļĦæĪIJåĬŁ
    -0.26
    ä¸ĢèĪ¬
    -0.25
    éné
    -0.25
    éĢ¢
    -0.24
     ping
    -0.24
    oris
    -0.24
    ä¸ĢèĪ¬çļĦ
    -0.24
    tit
    -0.24
    occasion
    -0.24
    ä¼ĺåħĪ
    -0.24
    POSITIVE LOGITS
    Subsystem
    0.29
    è¿ĻçĤ¹
    0.27
     hierarchy
    0.26
    æĽ²
    0.26
    $db
    0.25
     permissible
    0.24
    istine
    0.24
    åĩłä½ķ
    0.24
    管çIJĨç³»ç»Ł
    0.24
    elper
    0.24
    Activations Density 0.002%

    No Known Activations

    This feature has no known activations.