© 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. 129615
    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
    æĮĩ导æĦıè§ģ
    -0.28
    ouri
    -0.28
    .parentElement
    -0.25
    onacci
    -0.24
    éĽĨåĽ¢èĤ¡ä»½
    -0.24
    ainer
    -0.24
     toch
    -0.24
    isky
    -0.24
    驾
    -0.24
    Pri
    -0.24
    POSITIVE LOGITS
     cis
    0.27
    å½¢
    0.26
     CIS
    0.26
    æŁ³
    0.25
     sil
    0.25
     campaigns
    0.25
    ç¡ħ
    0.25
    Behavior
    0.25
    ...
    0.24
    线索
    0.24
    Activations Density 8.117%

    No Known Activations

    This feature has no known activations.