© 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. 57253
    Prev
    Next
    INDEX
    Explanations

    homeostasis, norm

    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
    ullo
    -0.12
    CACHE
    -0.10
    aps
    -0.10
    anco
    -0.09
    群ä¼Ĺ
    -0.09
    ç¾ŀ
    -0.09
    Messenger
    -0.09
    oks
    -0.09
    zos
    -0.09
    غÙĪ
    -0.09
    POSITIVE LOGITS
    liner
    0.10
    åĿĢ
    0.10
    isers
    0.09
    åģļæ³ķ
    0.09
    å±ĢéĿ¢
    0.09
    æŦ
    0.09
    öt
    0.09
    ä¹ĭä½ľ
    0.09
    -ish
    0.08
    çľ¼éĥ¨
    0.08
    Activations Density 0.020%

    No Known Activations