© 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. 51870
    Prev
    Next
    INDEX
    Explanations

    construction, railway, road, tunnel

    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
    æijĩ
    -0.10
    ÐľÐŀ
    -0.09
    ÑĤек
    -0.09
    пÑĥÑģÑĤ
    -0.09
    离
    -0.09
    Structured
    -0.09
     Bucc
    -0.08
    éĻ¤å¤ķ
    -0.08
    éĢ»
    -0.08
    _INITIALIZER
    -0.08
    POSITIVE LOGITS
    沿线
    0.11
    ä¸Ģæ¢Ŀ
    0.11
    ä¸ĢæĿ¡
    0.10
    è¿ĻæĿ¡
    0.10
    使人
    0.10
    æĽ²çº¿
    0.09
    éĢīåĿĢ
    0.09
    éļ§
    0.09
    éħ°
    0.09
    å¿łå®ŀ
    0.09
    Activations Density 0.067%

    No Known Activations