© 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. 32985
    Prev
    Next
    INDEX
    Explanations

    quantifying measurements

    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
    æĢ»çļĦ
    -0.10
    gay
    -0.09
    ä¼ļå±ķ
    -0.09
     أج
    -0.09
     Instant
    -0.09
    å®ĹæŨ
    -0.09
    ä½ľåĩºäºĨ
    -0.09
    çIJĨ念
    -0.08
    åģļå¾Ĺ
    -0.08
    ä»»æľŁ
    -0.08
    POSITIVE LOGITS
    æı£
    0.10
     *[
    0.09
    ä¸Ģç¯ĩ
    0.08
    ãģ¡ãĤī
    0.08
     pod
    0.08
    loo
    0.08
    çļĦä¸Ģ
    0.08
    çŃ
    0.08
    æīĺ
    0.08
    为åĩĨ
    0.08
    Activations Density 0.403%

    No Known Activations