© 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. 7454
    Prev
    Next
    INDEX
    Explanations

    non-English characters

    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
    åĦĺ
    -0.08
    iente
    -0.08
    \u
    -0.08
    aily
    -0.08
     Falk
    -0.08
    ane
    -0.08
    çī
    -0.08
     Lockheed
    -0.08
    /animate
    -0.08
    æļ§
    -0.07
    POSITIVE LOGITS
    ä¼ı
    0.11
    çħ³
    0.10
    æĻ¦
    0.09
    æĶ¾å®½
    0.09
    CHA
    0.09
    ç½ij讯
    0.09
    PHA
    0.08
    è¿Ļ个ä¸ĸçķĮ
    0.08
    è¿Ļä¸ĢçĤ¹
    0.08
    _PY
    0.08
    Activations Density 1.245%

    No Known Activations