© 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. 54870
    Prev
    Next
    INDEX
    Explanations

    secretions and discharge

    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
    éĩİçĶŁ
    -0.10
    è´¯
    -0.09
     erg
    -0.09
    ä¸Ģä½ĵåĮĸ
    -0.09
     alf
    -0.09
    ç´ł
    -0.09
    çł·
    -0.09
    èĸ°
    -0.08
    èĪĨ
    -0.08
     Settlement
    -0.08
    POSITIVE LOGITS
    ä»Ģä¹Īæł·çļĦ
    0.10
    çŃĴ
    0.10
     constructive
    0.09
    ìķħ
    0.09
    icap
    0.09
    å¤ļåįĬ
    0.09
    sey
    0.09
    ä¼´éļı
    0.09
    onder
    0.08
    enty
    0.08
    Activations Density 0.028%

    No Known Activations