© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52446
    Prev
    Next
    INDEX
    Explanations

    designer, pattern, making

    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
    oa
    -0.09
     deb
    -0.09
    æµ·è¾¹
    -0.09
    åľ°éľĩ
    -0.09
     åľ
    -0.08
    è¡Įãģ£ãģ¦
    -0.08
    iri
    -0.08
     Codec
    -0.08
    éĶ£
    -0.08
    Https
    -0.08
    POSITIVE LOGITS
    æĤ¨åIJĮæĦı
    0.09
    åħļçļĦåįģä¹Ŀ
    0.09
    éļIJèĶ½
    0.08
    èĬĻ
    0.08
    opor
    0.08
    ForResult
    0.08
    thro
    0.08
    æĤ¨çļĦ
    0.08
    人çļĦ
    0.08
    åĩĭ
    0.08
    Activations Density 0.076%

    No Known Activations