© 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. 27496
    Prev
    Next
    INDEX
    Explanations

    mild conditions and gentle requests

    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
     thrill
    -0.10
    æĬĹ
    -0.10
    åĭĩæ°Ķ
    -0.09
    ugin
    -0.09
    astes
    -0.09
    æĸĹå¿Ĺ
    -0.09
    éħĴ
    -0.08
    é£Łæ¬²
    -0.08
    åĬłåĽº
    -0.08
    éĹ»
    -0.08
    POSITIVE LOGITS
    温æŁĶ
    0.20
    æŁĶåĴĮ
    0.19
    温åĴĮ
    0.18
    æŁĶ
    0.17
     gent
    0.17
    对å¾ħ
    0.16
     gentle
    0.16
     Gentle
    0.14
    æŁĶ软
    0.14
    gent
    0.14
    Activations Density 0.069%

    No Known Activations