© 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. 3948
    Prev
    Next
    INDEX
    Explanations

    flat

    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
    éŀŃ
    -0.10
    å°ıå¿ĥ
    -0.09
    çīĿ
    -0.09
    ä»·æł¼ä¸Ĭ涨
    -0.09
    ettel
    -0.09
    (Column
    -0.09
    åįİ举
    -0.08
    .Protocol
    -0.08
    æīª
    -0.08
     Desmond
    -0.08
    POSITIVE LOGITS
    æīģ
    0.18
    -flat
    0.16
    .Flat
    0.16
    flat
    0.16
     flat
    0.16
    Flat
    0.15
    å¹³åĿ¦
    0.15
     Flat
    0.15
    .flat
    0.14
    _flat
    0.14
    Activations Density 0.069%

    No Known Activations