© 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. 7928
    Prev
    Next
    INDEX
    Explanations

    sucrose or sugar

    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
    诸èijĽ
    -0.09
    åѤ
    -0.09
    ç«łç¨ĭ
    -0.09
    å«ł
    -0.09
    é³ĸ
    -0.09
    æĦŁå®ĺ
    -0.09
    ä¸Ńè¶ħ
    -0.08
    å©Ĩå©Ĩ
    -0.08
    èĮĦ
    -0.08
    èĤīç±»
    -0.08
    POSITIVE LOGITS
    èĶĹ
    0.14
    ç³ĸ
    0.12
     caramel
    0.12
     ÑģаÑħаÑĢ
    0.11
     sugar
    0.10
    çĻ½ç³ĸ
    0.09
    ç³ĸå°¿çĹħ
    0.09
    äºķ
    0.09
     ÅŁeker
    0.09
     Sugar
    0.09
    Activations Density 0.117%

    No Known Activations