© 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. 22-TRANSCODER-HP
    7. 130166
    Prev
    Next
    INDEX
    Explanations

    asked

    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
    Not in Any Lists

    No Comments

    Negative Logits
    å¢ŀåĢ¼
    -0.24
    errs
    -0.24
    ANA
    -0.24
    ingu
    -0.24
     hearing
    -0.23
    .workflow
    -0.23
    çļĦæĦıæĢĿ
    -0.23
    ä»»èģĮ
    -0.23
    æ¥ļ
    -0.23
    district
    -0.23
    POSITIVE LOGITS
    åľ¨è¿Ļ
    0.28
    éĢĤæŶ
    0.27
    éĢĢ
    0.26
     faç
    0.25
    è¯Ŀ
    0.25
     Stat
    0.25
    æķ²
    0.24
    åĬ¡
    0.24
    rame
    0.24
    ç«Ł
    0.24
    Activations Density 0.024%

    No Known Activations