© 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. 98756
    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
     projector
    -0.27
     setLocation
    -0.26
     setText
    -0.24
    åĥļ
    -0.24
    (Constructor
    -0.24
    åħģ许
    -0.23
    ç»ıåİĨ
    -0.23
    é¡¹çĽ®å»ºè®¾
    -0.23
    .projects
    -0.22
    书记
    -0.22
    POSITIVE LOGITS
    اÙħا
    0.25
     incur
    0.24
    cis
    0.24
    Hi
    0.24
    ircon
    0.24
    åĩºè½¨
    0.23
    ä¸įåı¯
    0.23
    ivable
    0.22
    dar
    0.22
     easily
    0.22
    Activations Density 0.011%

    No Known Activations

    This feature has no known activations.