© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 20822
    Prev
    Next
    INDEX
    Explanations

    possessive pronouns followed by locations or possessions

    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
    髯
    -0.10
    è³ŀ
    -0.10
     haired
    -0.09
    çļĦåij³éģĵ
    -0.09
    åıijæī¬
    -0.09
    å°±æĺ¯ä¸Ģ个
    -0.08
    æĿµ
    -0.08
     traf
    -0.08
    -haired
    -0.08
     gradual
    -0.08
    POSITIVE LOGITS
    æīĢåľ¨çļĦ
    0.12
    ä¸ĵå±ŀ
    0.10
    çļĦçĶŁæ´»
    0.09
    çĶŁæ´»çݯå¢ĥ
    0.09
    èĢĮéĿŀ
    0.09
    è¹Ń
    0.08
    /her
    0.08
    æĹ¥å¸¸çĶŁæ´»
    0.08
     home
    0.08
    ä½ıæīĢ
    0.08
    Activations Density 0.089%

    No Known Activations