© 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. 130920
    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
    pants
    -0.32
     AJ
    -0.27
    <main
    -0.26
    cloth
    -0.25
    hon
    -0.25
    æ´ĭæ´ĭ
    -0.24
    hoff
    -0.24
    rets
    -0.24
    elly
    -0.24
    hone
    -0.24
    POSITIVE LOGITS
    rze
    0.28
    ä¿¡æģ¯æľįåĬ¡
    0.26
    羡æħķ
    0.26
     benchmark
    0.25
     earning
    0.24
    ue
    0.24
     libr
    0.24
     experiment
    0.24
    library
    0.23
     quant
    0.23
    Activations Density 4.916%

    No Known Activations

    This feature has no known activations.