© 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. 21242
    Prev
    Next
    INDEX
    Explanations

    continuous or breaking patterns

    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
    æĸĻ
    -0.10
     Closet
    -0.09
    yas
    -0.09
    æĹĮ
    -0.08
     consolid
    -0.08
    ãĥŃãĥ³
    -0.08
     AMP
    -0.08
    çºł
    -0.08
    éħ°
    -0.08
    emma
    -0.08
    POSITIVE LOGITS
    è¿ĩ渡
    0.17
    è¡Ķæİ¥
    0.14
    æ¸IJ
    0.13
     пеÑĢеÑħод
    0.12
     zone
    0.12
    æŃ¢
    0.12
     abrupt
    0.11
    éĺ¶æ¢¯
    0.11
    åľ°å¸¦
    0.11
    交æİ¥
    0.11
    Activations Density 0.193%

    No Known Activations