© 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. 2862
    Prev
    Next
    INDEX
    Explanations

    plasticizers, excipients, base, mineral, silicone, clay, potassium

    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
    emo
    -0.10
    客åİħ
    -0.09
    åįĬ个
    -0.09
    ಲ
    -0.09
    vu
    -0.09
    oad
    -0.08
    æł¸å¿ĥåĮº
    -0.08
    设æĥ³
    -0.08
    ynes
    -0.08
    aris
    -0.08
    POSITIVE LOGITS
    诸
    0.10
    çĸı
    0.09
    é£İçŃĿ
    0.09
    å±ıèĶ½
    0.09
    ulan
    0.09
    çľ¼ç§ij
    0.09
    绩
    0.09
    æĭ±
    0.09
    æµİ
    0.09
    ä¾į
    0.09
    Activations Density 0.121%

    No Known Activations