© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52003
    Prev
    Next
    INDEX
    Explanations

    midges or bugs

    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
    æħİ
    -0.09
    ibr
    -0.09
    éĶģå®ļ
    -0.09
    æľ¬æĿ¥
    -0.09
    pai
    -0.09
    é²ĩ
    -0.09
    çīĮåŃIJ
    -0.08
    ç¾ģ
    -0.08
    é¦Ĩ
    -0.08
    访
    -0.08
    POSITIVE LOGITS
    没è§ģè¿ĩ
    0.11
    ENDOR
    0.10
    å°¾
    0.09
    æĭ§
    0.09
    åĽ½éĻħæľºåľº
    0.09
    _DIP
    0.09
    踮
    0.09
    swith
    0.09
    æĪ¬
    0.09
    éŃĶæľ¯
    0.08
    Activations Density 0.093%

    No Known Activations