© 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. 130085
    Prev
    Next
    INDEX
    Explanations

    -

    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
     grade
    -0.30
    å°ijå¹´
    -0.29
     Grade
    -0.29
    åIJijå¾Ģ
    -0.28
    åĪĨå¼Ģ
    -0.28
    ling
    -0.27
    agog
    -0.27
     adolescence
    -0.26
     Schneider
    -0.26
    ç§ĺå¯Ĩ
    -0.25
    POSITIVE LOGITS
    nal
    0.30
     ranges
    0.28
    饮éħĴ
    0.28
    iola
    0.27
     cryptoc
    0.27
     jur
    0.26
    ç£ĭ
    0.26
     currentDate
    0.25
     nal
    0.25
    respond
    0.25
    Activations Density 0.055%

    No Known Activations