© 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. 2427
    Prev
    Next
    INDEX
    Explanations

    pod spec type

    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
    åĽĽåij¨
    -0.10
    éŁ³ä¹IJä¼ļ
    -0.10
    åѤç«ĭ
    -0.09
    ä¸įå°ı
    -0.09
    ä¸ĩä¸Ģ
    -0.09
    zion
    -0.09
    rze
    -0.09
    çĺķ
    -0.09
    esian
    -0.09
    ffective
    -0.09
    POSITIVE LOGITS
    icot
    0.10
    èĬĤ
    0.09
    ling
    0.09
    ç¯Ģ
    0.09
    ags
    0.09
     jes
    0.08
    åĿļ飧
    0.08
    quires
    0.08
    -outs
    0.08
    imately
    0.08
    Activations Density 0.053%

    No Known Activations