© 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. 37227
    Prev
    Next
    INDEX
    Explanations

    relational model

    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
    訴
    -0.09
    %x
    -0.09
    山寨
    -0.09
    amel
    -0.09
    Jam
    -0.08
    æ¤į
    -0.08
    æ®ĸ
    -0.08
    èĻ¢
    -0.08
    æ¶Īè´¹èĢħçļĦ
    -0.08
    æ±ĩ
    -0.08
    POSITIVE LOGITS
    é»İæĺİ
    0.10
     Dems
    0.09
     Laud
    0.09
    union
    0.09
    å½ķ
    0.09
    åĩºç¤º
    0.09
    oze
    0.08
     вÑĭполнен
    0.08
    å¦ĤæĦı
    0.08
    å®Įæķ´æĢ§
    0.08
    Activations Density 0.179%

    No Known Activations