© 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. 35235
    Prev
    Next
    INDEX
    Explanations

    frost, icing, ice

    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
    éļ¾å¾Ĺ
    -0.09
    æµ´
    -0.09
    æ²īé»ĺ
    -0.09
    èIJ½åı¶
    -0.09
    velt
    -0.09
    jin
    -0.08
    åijķ
    -0.08
    çĭ©
    -0.08
     Promo
    -0.08
    大æ¦Ĥ
    -0.08
    POSITIVE LOGITS
    åįĥæĸ¹çĻ¾è®¡
    0.10
    åħ¬æĬ¥
    0.10
    åģľçķĻåľ¨
    0.09
    å½ķçĶ¨
    0.09
    对ä¼ģä¸ļ
    0.09
    éľĪ
    0.09
    éĿł
    0.09
     Burl
    0.08
    OrUpdate
    0.08
    æ¶Īéĺ²å®īåħ¨
    0.08
    Activations Density 0.019%

    No Known Activations