© 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. 24032
    Prev
    Next
    INDEX
    Explanations

    .com

    np_max-act · gemini-2.0-flash
    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
     mı
    -0.28
    æĪijä¸įæĥ³
    -0.27
    æİĮæı¡
    -0.26
    她说
    -0.26
    èĨĺ
    -0.25
     lá»Ļ
    -0.24
    semblies
    -0.24
    acies
    -0.23
    UCT
    -0.23
    enance
    -0.23
    POSITIVE LOGITS
    ĪæĿĥ
    0.32
    æĪIJéķ¿ä¸º
    0.27
    ahoma
    0.27
    Their
    0.24
    è¯ķåį·
    0.24
    ets
    0.24
     their
    0.24
    ford
    0.24
    çªģçł´
    0.24
     THEIR
    0.23
    Activations Density 0.002%

    No Known Activations