© 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. 108805
    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
    缶
    -0.27
     cabe
    -0.26
    urement
    -0.26
    ustral
    -0.26
     tariff
    -0.26
     attracted
    -0.25
     mutually
    -0.25
     mutual
    -0.25
    æŃĮ
    -0.25
    åŃ
    -0.25
    POSITIVE LOGITS
    è¿Ļ件äºĭæĥħ
    0.27
    abcd
    0.27
    ABI
    0.26
    è¿Ļå¥Ĺ
    0.25
     Nicolas
    0.24
    ä¼ļå½±åĵį
    0.24
    æģIJæĢķ
    0.24
    <select
    0.24
    (G
    0.24
    åĬłä»¥
    0.24
    Activations Density 1.763%

    No Known Activations

    This feature has no known activations.