© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 20-TRANSCODER-HP
    7. 142708
    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
    Negative Logits
    \č↵
    -0.30
    ktop
    -0.29
    onation
    -0.27
    è·¨è¶Ĭ
    -0.24
    jÄĻ
    -0.24
    .ADD
    -0.24
     Modal
    -0.24
    ymoon
    -0.24
    пÑĤ
    -0.23
    ulpt
    -0.23
    POSITIVE LOGITS
    gend
    0.30
     supplies
    0.25
    åºĨ
    0.25
     Seite
    0.24
     freshly
    0.24
     Raleigh
    0.24
    .forChild
    0.24
     saÄŁ
    0.24
    åı£æ°´
    0.24
    Configurer
    0.23
    Activations Density 36.395%

    No Known Activations

    This feature has no known activations.