© 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. 63253
    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
     tuition
    -0.28
    adows
    -0.26
    tutorial
    -0.26
    oppins
    -0.24
    bits
    -0.24
    codegen
    -0.24
    ornings
    -0.24
     erv
    -0.24
    pectives
    -0.23
    ogui
    -0.23
    POSITIVE LOGITS
    ovel
    0.28
    ä¸įå¼Ģ
    0.27
     fin
    0.27
    ns
    0.26
     rico
    0.26
     ri
    0.25
    ä¸Ģç«Ļ
    0.25
    åIJĦè¡Į
    0.25
    'n
    0.24
    yna
    0.24
    Activations Density 49.666%

    No Known Activations

    This feature has no known activations.