© 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. 31538
    Prev
    Next
    INDEX
    Explanations

    possessive 's followed by disease

    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
    è¾Ī
    -0.10
    èĢģ家
    -0.10
    less
    -0.10
    ä¼ĹæīĢ
    -0.09
    overs
    -0.09
    ors
    -0.09
    ä¸¤åĽ½
    -0.09
    æīĭä¸ĭ
    -0.09
    æĪIJéķ¿
    -0.09
    sts
    -0.09
    POSITIVE LOGITS
    ฯ
    0.09
    æľ¬æľŁ
    0.09
    erg
    0.08
    ä¸Ģ页
    0.08
    ìĺ¥
    0.08
    YSTEM
    0.08
    åĪĽéĢłäºĨ
    0.08
    æļĹæļĹ
    0.08
    ourke
    0.08
    cales
    0.08
    Activations Density 0.028%

    No Known Activations