© 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. 12-TRANSCODER-HP
    7. 24137
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_12.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.12.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    æĽ´å¼º
    -0.27
    spot
    -0.26
     WP
    -0.25
     street
    -0.25
     responsible
    -0.25
    пÑĥÑģка
    -0.25
     numer
    -0.25
     weak
    -0.25
    æĹłéĻIJ
    -0.25
    èģĮåľº
    -0.25
    POSITIVE LOGITS
    DUCT
    0.30
    å°º
    0.28
    æĹı
    0.27
    isin
    0.27
    çĮĬ
    0.25
    _HC
    0.24
    .plugins
    0.24
    ÐIJÐŀ
    0.23
    bw
    0.23
     Surgery
    0.23
    Activations Density 0.002%

    No Known Activations

    This feature has no known activations.