© 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. 108445
    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
    ViewById
    -0.31
    icas
    -0.28
    iche
    -0.26
    è£ı
    -0.25
    acker
    -0.25
    课åłĤæķĻåѦ
    -0.24
     belly
    -0.24
    osto
    -0.24
    å½ķ
    -0.24
    çļĦçĪ¶äº²
    -0.24
    POSITIVE LOGITS
    ture
    0.30
    bury
    0.27
    hog
    0.26
    çģ
    0.25
    \v
    0.24
    unread
    0.24
    .invoke
    0.24
    éĢļè¿ĩ
    0.24
    循
    0.23
     ['-
    0.23
    Activations Density 2.446%

    No Known Activations

    This feature has no known activations.