© 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. 20-TRANSCODER-HP
    7. 130642
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_20.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.20.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    工夫
    -0.28
    .bel
    -0.27
    via
    -0.26
    _managed
    -0.26
    anja
    -0.25
    æķ´é¡¿
    -0.25
     Forms
    -0.24
     via
    -0.23
    å½ĵå±Ģ
    -0.23
    gage
    -0.23
    POSITIVE LOGITS
    å°Ī
    0.25
     sĩ
    0.25
    utas
    0.25
    eto
    0.25
    å°Ĩè¿ij
    0.25
    æµģè¡Į
    0.25
    ottes
    0.24
    ót
    0.24
    ä¸ĭ游
    0.24
     isolate
    0.24
    Activations Density 12.432%

    No Known Activations

    This feature has no known activations.