© 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. 20191
    Prev
    Next
    INDEX
    Explanations

    from a

    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
    åIJ¸æĶ¶
    -0.10
    ç»¿åľ°
    -0.08
    akra
    -0.08
    çĶ¸
    -0.08
     Voice
    -0.08
    æĬĵä½ı
    -0.08
    .getB
    -0.08
     bomb
    -0.08
    olv
    -0.08
     Homepage
    -0.08
    POSITIVE LOGITS
    à¸Ľà¸£à¸²
    0.10
    äºĨä¸Ģåı£
    0.10
    èĥ¯
    0.09
    çħŀ
    0.09
    ADS
    0.08
    çķħéĢļ
    0.08
    pei
    0.08
    ëįķ
    0.08
    éŨæĪ·
    0.08
    é«ĺæĺĤ
    0.08
    Activations Density 0.056%

    No Known Activations