© 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. 23653
    Prev
    Next
    INDEX
    Explanations

    of 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
    缸äºĴ
    -0.08
    çĶ©
    -0.08
    æĶ¾æīĭ
    -0.08
    殳
    -0.08
     Thousand
    -0.08
    çģ¶
    -0.08
    çĦĺ
    -0.08
    ep
    -0.08
    æijĩ
    -0.08
    æķ´æķ´
    -0.08
    POSITIVE LOGITS
    éĵ®
    0.10
     tán
    0.09
    èİ«æĸ¯
    0.09
    illions
    0.09
    èĿĩ
    0.09
    åŀŁ
    0.09
    ä¹Łå¥½
    0.09
    öst
    0.08
    _usec
    0.08
    çIJ¯
    0.08
    Activations Density 0.023%

    No Known Activations