© 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. 107289
    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
    ابة
    -0.29
     eligible
    -0.25
    .setAction
    -0.25
     Pare
    -0.25
    udios
    -0.24
    ’
    -0.24
     chunks
    -0.24
    odge
    -0.24
     advant
    -0.24
     skeletons
    -0.24
    POSITIVE LOGITS
    主ä¹īæĢĿæĥ³
    0.27
    enus
    0.25
    ç½ijç»ľä¼łæĴŃ
    0.25
    stras
    0.25
    æİ¨è¿Ľ
    0.25
    thr
    0.25
    meeting
    0.24
    omm
    0.24
    oom
    0.24
    _ser
    0.24
    Activations Density 4.185%

    No Known Activations

    This feature has no known activations.