© 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. 141563
    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
    qui
    -0.26
    venience
    -0.26
    venient
    -0.26
    enna
    -0.25
    apg
    -0.25
    ä¸ĵ人
    -0.25
    æłĸ
    -0.24
     cerv
    -0.24
     Electoral
    -0.24
    åıĸèĥľ
    -0.24
    POSITIVE LOGITS
    è¾ij
    0.29
    inci
    0.28
    IJľç´¢
    0.27
    æĭ§
    0.25
    OutputStream
    0.25
    æĸ°ä¸Ģ代
    0.25
    nung
    0.25
    INC
    0.25
     Prep
    0.24
    ç»ĵ
    0.24
    Activations Density 6.058%

    No Known Activations

    This feature has no known activations.