© 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. 107967
    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
    æݬ
    -0.30
    Ø´Ùģ
    -0.27
     slur
    -0.27
    è¾Ł
    -0.25
    好å¤Ħ
    -0.25
     ÙĨÙĬÙĪز
    -0.24
    inos
    -0.24
    åŃº
    -0.24
    å¸Ń
    -0.23
     skinny
    -0.23
    POSITIVE LOGITS
    æ³°å±±
    0.30
    ende
    0.25
    оÑģÑĤÑĮÑİ
    0.25
    ucion
    0.24
    é¹³
    0.24
    å±±ä¸Ĭ
    0.24
    PTH
    0.24
    çī¹éķ¿
    0.24
    agements
    0.24
     tem
    0.23
    Activations Density 1.282%

    No Known Activations

    This feature has no known activations.