© 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. 33668
    Prev
    Next
    INDEX
    Explanations

    esophagitis, allergy, contaminated foods, barrier impairment, stalking, Qt, proper nouns

    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
    æį¨ãģ¦
    -0.10
    hani
    -0.09
    çĻ¾ç§ij
    -0.09
    ä¸ĥåįģ
    -0.09
    çģ«çĦ°
    -0.09
    ç¯ĵ
    -0.09
    ÑĤов
    -0.09
    æĭĶ
    -0.09
    çĽĨ
    -0.09
    åľ¨äº¬
    -0.08
    POSITIVE LOGITS
     allerg
    0.12
    é²²
    0.11
    é³Į
    0.11
    è¿ĩæķı
    0.10
    å®īæ£Ģ
    0.09
    鲤
    0.09
    èł²
    0.09
    é²İ
    0.09
    foods
    0.09
    è§īå¾Ĺ
    0.09
    Activations Density 0.068%

    No Known Activations