© 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. 32-TRANSCODER-HP
    7. 42968
    Prev
    Next
    INDEX
    Explanations

    say "vision"

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    mwhanna/qwen3-4b-transcoders/layer_32.safetensors
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    163,840
    Data Type
    float32
    Hook Name
    blocks.32.mlp.hook_in
    Architecture
    transcoder
    Context Size
    8,192
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ellar
    -0.26
    æµĥ
    -0.26
    顧
    -0.26
    纤维
    -0.25
    zyÄĩ
    -0.25
    kaar
    -0.25
    ä¸į说
    -0.25
    çŃ¼
    -0.25
    累累
    -0.25
    record
    -0.25
    POSITIVE LOGITS
    emption
    0.29
     abandonment
    0.27
    uhl
    0.27
    ãģ¡ãģ¯
    0.26
    空
    0.25
    oons
    0.25
    acios
    0.25
    atal
    0.24
     development
    0.24
     volatility
    0.24
    Activations Density 0.002%

    No Known Activations