© 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. 142753
    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
    qui
    -0.30
    vt
    -0.29
    ÅŃ
    -0.28
    histor
    -0.27
    yst
    -0.26
     dyn
    -0.26
    estro
    -0.25
    山谷
    -0.25
    afc
    -0.25
    è¯Ħä¼°
    -0.25
    POSITIVE LOGITS
    edith
    0.27
    åĩŃä»Ģä¹Ī
    0.26
    ikers
    0.26
    eways
    0.26
    æĭĽåĭŁ
    0.25
    ãģľ
    0.25
    ä¸įå¦Ĥ
    0.25
     ims
    0.25
    kok
    0.25
    okus
    0.25
    Activations Density 5.609%

    No Known Activations

    This feature has no known activations.