© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-4B
    5. 20-TRANSCODER-HP
    7. 107991
    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
    omat
    -0.27
    æݨèįIJéĺħ读
    -0.26
    çĽĺæ´»
    -0.26
    mers
    -0.25
    esty
    -0.25
    Format
    -0.24
     neighbouring
    -0.24
    chematic
    -0.24
    _um
    -0.23
     Format
    -0.23
    POSITIVE LOGITS
     except
    0.29
    ä»Ļ
    0.28
    æ±Ń
    0.28
    åıĺ
    0.28
     alike
    0.27
    Routine
    0.27
    æĪijå¿ĥéĩĮ
    0.27
    åģıå·®
    0.27
     départ
    0.26
    bai
    0.25
    Activations Density 51.459%

    No Known Activations

    This feature has no known activations.