© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 39475
    Prev
    Next
    INDEX
    Explanations

    team's or teams have

    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
    éĢIJå¹´
    -0.10
    elines
    -0.10
    èĢģ人家
    -0.10
    ahren
    -0.09
    gers
    -0.09
    çĪª
    -0.09
     ÐľÐ°
    -0.09
    åĩºçİ°åľ¨
    -0.09
    èħł
    -0.09
    endor
    -0.09
    POSITIVE LOGITS
    ç¥ļ
    0.10
    çĴĭ
    0.08
    unts
    0.08
    rab
    0.08
    常çĶ¨
    0.08
    -wide
    0.08
    offs
    0.08
     hin
    0.08
    _shot
    0.08
    woke
    0.08
    Activations Density 0.008%

    No Known Activations