© 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. 64864
    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
    hana
    -0.27
    èĬ®
    -0.25
     trim
    -0.24
    滩
    -0.23
    haft
    -0.23
    å¼ķãģį
    -0.23
    otify
    -0.23
    ryption
    -0.23
    [src
    -0.23
     resh
    -0.23
    POSITIVE LOGITS
    åŃĺåľ¨çļĦ
    0.27
    ç»ĵ论
    0.26
    å¦ĤæѤ
    0.26
     Ny
    0.25
    æĮ¥
    0.25
    ypo
    0.24
    çļĦåľ°
    0.24
    åľ¨åĽ½å¤ĸ
    0.24
    weg
    0.24
    Contrib
    0.24
    Activations Density 2.348%

    No Known Activations

    This feature has no known activations.