© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 49449
    Prev
    Next
    INDEX
    Explanations

    everyone followed by verbs

    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
    ä¸įæĺĵ
    -0.10
    并ä¸įå¤ļ
    -0.09
    agoon
    -0.09
    å®Įæķ´
    -0.09
    ular
    -0.09
    æĩ¿
    -0.09
    ä¸į太好
    -0.09
    å·¡
    -0.08
    并æĹł
    -0.08
    表æī¬
    -0.08
    POSITIVE LOGITS
    éĥ½çŁ¥éģĵ
    0.11
     вокÑĢÑĥг
    0.11
    人人éĥ½
    0.11
    æĪij们éĥ½
    0.11
    except
    0.10
    人人
    0.10
    åĽ´ç»ķ
    0.10
    éĥ½åľ¨
    0.10
    igator
    0.10
    éĥ½è¯´
    0.10
    Activations Density 0.048%

    No Known Activations