© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 3923
    Prev
    Next
    INDEX
    Explanations

    gazing and looking

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    æľīæĿ¡ä»¶çļĦ
    -0.10
    è§Ĥ念
    -0.09
    ä»Ģ麽
    -0.08
    мÑĭ
    -0.08
    teen
    -0.08
    太åIJİ
    -0.08
    gy
    -0.08
     tÆ°
    -0.07
    åĩŃçĿĢ
    -0.07
     sound
    -0.07
    POSITIVE LOGITS
    -eyed
    0.13
    ãģĤãĤĮ
    0.11
    ight
    0.10
    çļĦ缮åħī
    0.10
    еÑĤÑĮ
    0.10
    æľĽåİ»
    0.10
    çªĹå¤ĸ
    0.10
    argout
    0.09
    çľ¼
    0.09
    عÙĬÙĨ
    0.09
    Activations Density 0.005%

    No Known Activations