© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 50783
    Prev
    Next
    INDEX
    Explanations

    special punctuation

    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
    est
    -0.13
    ties
    -0.11
    ments
    -0.10
    halb
    -0.10
    -être
    -0.10
    èĢģ人家
    -0.10
    -thirds
    -0.10
    zelf
    -0.10
    nings
    -0.09
    åIJĦæł·
    -0.09
    POSITIVE LOGITS
    ãģĦãģij
    0.11
    اش
    0.11
    æł·çļĦ
    0.10
    ãģŁãģı
    0.10
    ãģĤãĤĭç¨ĭ度
    0.10
    ا
    0.10
    itude
    0.09
     doom
    0.09
    ëģĶ
    0.09
    ../../../
    0.09
    Activations Density 0.209%

    No Known Activations