© 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. 98685
    Prev
    Next
    INDEX
    Explanations

    Guinea/inea/GNU

    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
    ed
    -0.38
    ynes
    -0.28
    ced
    -0.25
    dotenv
    -0.25
    eded
    -0.25
    dır
    -0.25
    aped
    -0.24
    ä¸Ģ个å¤ļ
    -0.23
    icina
    -0.23
    æµģéĢĿ
    -0.23
    POSITIVE LOGITS
    èĤł
    0.29
     happily
    0.25
     onRequest
    0.25
    é¼ĭ
    0.24
    earch
    0.24
    Toast
    0.23
    è§Ĥ
    0.23
     ounce
    0.23
    à¹īำ
    0.23
     secretary
    0.23
    Activations Density 0.054%

    No Known Activations