© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 50784
    Prev
    Next
    INDEX
    Explanations

    end user or end application

    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
    强æľīåĬĽçļĦ
    -0.10
     Tos
    -0.09
    ongan
    -0.08
    ç§ijæĬĢèĤ¡ä»½
    -0.08
    ç©°
    -0.08
    æ·±æĥħ
    -0.08
    ä»İæł¹æľ¬
    -0.08
     surprise
    -0.08
    opers
    -0.08
     TSR
    -0.08
    POSITIVE LOGITS
    å¤Ł
    0.09
    åĴ¿
    0.09
    endif
    0.08
     дело
    0.08
    à¹īà¸Ńà¸Ļ
    0.08
    ugen
    0.08
    ért
    0.08
    çĭİ
    0.08
     thiá»ĩn
    0.08
    ä¸įä½ı
    0.08
    Activations Density 0.101%

    No Known Activations