© 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. 6903
    Prev
    Next
    INDEX
    Explanations

    Cyrillic characters related to 'заказ' and 'роз'

    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
    astr
    -0.11
    ellant
    -0.10
    reme
    -0.09
    otine
    -0.09
    åģĩæĹ¥
    -0.09
    ä¼´æľī
    -0.08
    OPLE
    -0.08
     Ot
    -0.08
     Gang
    -0.08
     Edu
    -0.08
    POSITIVE LOGITS
    rous
    0.10
    æĮ¯åħ´
    0.10
    ially
    0.10
    мÑĭÑĪ
    0.09
    ending
    0.09
    çŀij
    0.09
    个å¤ļ
    0.09
    iates
    0.09
    UILT
    0.09
    iate
    0.09
    Activations Density 0.024%

    No Known Activations