© 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. 9386
    Prev
    Next
    INDEX
    Explanations

    highlight of

    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
    æĹĮ
    -0.09
    çļĦéĩįè¦ģæĢ§
    -0.09
     ба
    -0.09
    UIS
    -0.09
    åģļä¸įåĪ°
    -0.08
    asco
    -0.08
    shaw
    -0.08
    modo
    -0.08
    trigger
    -0.08
    æij©
    -0.08
    POSITIVE LOGITS
    æīĢåľ¨
    0.12
    亮çĤ¹
    0.11
    ä¸įä¹ı
    0.10
    lett
    0.10
    abol
    0.10
    许å¤ļ人
    0.09
    æĹłæķ°
    0.09
    èİ«è¿ĩäºİ
    0.09
     cul
    0.08
    è¡Įç¨ĭ
    0.08
    Activations Density 0.019%

    No Known Activations