© 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. 2692
    Prev
    Next
    INDEX
    Explanations

    figured or figuring

    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
    æĪijè§īå¾Ĺ
    -0.12
    èĩ³äºİ
    -0.11
    æģ³
    -0.11
    æĪij认为
    -0.10
    oto
    -0.10
    èĩ³æĸ¼
    -0.09
    åĹŁ
    -0.09
    æł¹æľ¬æ²¡æľī
    -0.09
    羣çļĦå¾Ī
    -0.09
    ported
    -0.09
    POSITIVE LOGITS
    RESS
    0.09
    ulus
    0.09
    uss
    0.09
    ress
    0.08
    ton
    0.08
    å»ĵ
    0.08
    äºĨä¸Ģä¸ĭ
    0.08
    ãģ¹ãģį
    0.08
    Authorities
    0.08
     out
    0.08
    Activations Density 0.151%

    No Known Activations