© 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. 108323
    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
    ament
    -0.29
    pez
    -0.29
    ilk
    -0.28
     stages
    -0.27
    mente
    -0.27
    ments
    -0.26
    uellement
    -0.26
    iren
    -0.26
    ণ
    -0.26
    å¦Ĥæŀľæľī
    -0.24
    POSITIVE LOGITS
     rall
    0.26
    arl
    0.25
    обав
    0.25
     pumped
    0.25
    andscape
    0.24
     Unblock
    0.24
    upyter
    0.24
    æŀ¶
    0.24
     GLsizei
    0.23
     Wahl
    0.23
    Activations Density 34.800%

    No Known Activations

    This feature has no known activations.