© 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. 8228
    Prev
    Next
    INDEX
    Explanations

    curry Mandarin scrambled fry

    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
    urat
    -0.09
    Ùĥر
    -0.09
    åŃĹåı·
    -0.09
    íŀĪ
    -0.09
    ijken
    -0.08
    IH
    -0.08
    æ¿ł
    -0.08
    IQ
    -0.08
    icens
    -0.08
    пÑĢом
    -0.08
    POSITIVE LOGITS
    hana
    0.10
    缨
    0.10
    快讯
    0.10
    樾
    0.09
    lund
    0.09
    kek
    0.09
    еÑĢж
    0.09
    gether
    0.09
    篷
    0.09
    _pdu
    0.08
    Activations Density 0.006%

    No Known Activations