© 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. 36678
    Prev
    Next
    INDEX
    Explanations

    Proguard, unpack, minify, jadx

    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
    LU
    -0.10
    ran
    -0.09
    å¦Ĥå®ŀ
    -0.08
    aires
    -0.08
    åı£ä¸Ń
    -0.08
     paper
    -0.08
     audi
    -0.08
    çŃ¹
    -0.08
    orsche
    -0.08
    就说
    -0.08
    POSITIVE LOGITS
    /goto
    0.10
    é¼İ
    0.10
    éľĵ
    0.10
    åī¥
    0.10
    çľĭä¸įæĩĤ
    0.09
     мн
    0.09
    нев
    0.09
    éĢĨ
    0.09
    É©
    0.09
    è¾ij
    0.09
    Activations Density 0.298%

    No Known Activations