© 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. 98102
    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
    Not in Any Lists

    No Comments

    Negative Logits
     :↵↵↵↵
    -0.26
    é²ľèī³
    -0.24
     :]
    -0.24
     DIE
    -0.23
     optic
    -0.22
     carved
    -0.22
     certain
    -0.21
    PRIVATE
    -0.21
    ä¹Łä¸įä¾ĭå¤ĸ
    -0.21
    èī³
    -0.21
    POSITIVE LOGITS
     Opposition
    0.25
    æľ¬æĿ¥
    0.24
    Opens
    0.23
    æīĵå¼ĢäºĨ
    0.23
    TX
    0.23
    湫
    0.23
    ÄĻd
    0.23
     TX
    0.22
    çļĦ设计
    0.22
     fool
    0.21
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.