© 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. 112645
    Prev
    Next
    INDEX
    Explanations

    materials

    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
    rades
    -0.28
    çļĦä¾ĭåŃIJ
    -0.27
     %=
    -0.27
    rade
    -0.26
    ERIC
    -0.26
    rhs
    -0.26
    é¢ĺ主
    -0.25
    楼主
    -0.25
     cap
    -0.25
    eric
    -0.25
    POSITIVE LOGITS
     Made
    0.24
    ä¸Ģ个å¤ļ
    0.23
    rove
    0.23
     gala
    0.23
    èIJ¥åķĨçݯå¢ĥ
    0.23
    .performance
    0.23
    éĺĸ
    0.23
    ournament
    0.23
     ("<
    0.22
    人éĥ½
    0.22
    Activations Density 0.008%

    No Known Activations