© 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. 58982
    Prev
    Next
    INDEX
    Explanations

    of their, in the

    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
     ÑģможеÑĤе
    -0.10
    è¿Ļæĺ¯æĪij们
    -0.10
    è¿ĻäºĽäºº
    -0.10
    æĪijè¦ģ
    -0.09
    _Public
    -0.09
    è¿Ļæĺ¯
    -0.09
    çľĭä½ł
    -0.08
    涯
    -0.08
    æıŃ
    -0.08
    è¿ĺæĺ¯è¦ģ
    -0.08
    POSITIVE LOGITS
    å®ĥçļĦ
    0.30
     its
    0.28
     Its
    0.23
    Its
    0.22
    åħ¶
    0.21
    使åħ¶
    0.19
    its
    0.19
    让å®ĥ
    0.18
    åĽłä¸ºå®ĥ
    0.17
    为åħ¶
    0.17
    Activations Density 0.710%

    No Known Activations