© 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. 108345
    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
     unst
    -0.27
    onga
    -0.26
    IEWS
    -0.26
    -automatic
    -0.25
    人éĥ½
    -0.25
    pañ
    -0.25
    ä¼ļç»Ļ
    -0.25
    —"
    -0.24
    ãģ¹ãģį
    -0.23
    åįļæĸĩ
    -0.23
    POSITIVE LOGITS
    çĨµ
    0.27
    odont
    0.25
    GNU
    0.25
     gấp
    0.25
    SU
    0.24
    buster
    0.24
     Finals
    0.24
    æĹ¢
    0.24
    antine
    0.24
    od
    0.24
    Activations Density 42.059%

    No Known Activations

    This feature has no known activations.