© 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. 8230
    Prev
    Next
    INDEX
    Explanations

    medium followed by range, term, or body

    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
     mere
    -0.10
    iform
    -0.10
    _mirror
    -0.09
    é¡ĺ
    -0.09
    åĸµ
    -0.09
    geber
    -0.08
    èĽ¸
    -0.08
    isphere
    -0.08
    iedad
    -0.08
    ÑĩеÑĤ
    -0.08
    POSITIVE LOGITS
    -sized
    0.18
     sized
    0.12
    /he
    0.10
    æĥħåĨµä¸ĭ
    0.10
    est
    0.10
    é©·
    0.09
     à¹Ĩ
    0.09
    ieval
    0.09
    à¹Ĩ
    0.09
    ร
    0.09
    Activations Density 0.013%

    No Known Activations