© 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. 9091
    Prev
    Next
    INDEX
    Explanations

    data type classification

    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
    çĶŁæ´»æĸ¹å¼ı
    -0.09
    è¶ĭ
    -0.09
     taps
    -0.09
    ç»Ħç»ĩé¢Ĩ导
    -0.09
    é«ĺå³°
    -0.08
    å«Į
    -0.08
    åıijå±ķ模å¼ı
    -0.08
    åĬ³ç´¯
    -0.08
    æĸ½èĤ¥
    -0.08
    ords
    -0.08
    POSITIVE LOGITS
    ç§įç±»
    0.10
    ä»Ģä¹Īéĥ½
    0.10
    ç±»åŀĭ
    0.09
    æĹ¢åı¯ä»¥
    0.09
    escal
    0.09
    丸
    0.09
    åIJģ
    0.09
    ensitive
    0.09
     types
    0.09
    ŀĭ
    0.09
    Activations Density 0.572%

    No Known Activations