© 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. 22-TRANSCODER-HP
    7. 131106
    Prev
    Next
    INDEX
    Explanations

    ulator

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_22.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.22.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Negative Logits
    ELL
    -0.29
    виÑģ
    -0.27
    еÑģÑĤи
    -0.26
    éĽ²
    -0.26
    roperty
    -0.25
    æ¹ĸåĮº
    -0.25
    åĿıäºĨ
    -0.25
    айÑĤе
    -0.25
    hell
    -0.24
    æĬķåħ¥åĪ°
    -0.24
    POSITIVE LOGITS
    ativa
    0.26
    åĽ¾æĸĩ
    0.24
    åĽ´
    0.24
     crust
    0.24
    ãģľãģ²
    0.24
     offended
    0.23
    æĭ³
    0.23
    è¿ŀ
    0.23
    atives
    0.23
    çĶŁåŃĺ
    0.23
    Activations Density 0.001%

    No Known Activations