© 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. 11146
    Prev
    Next
    INDEX
    Explanations

    -ful, -lessly, -ivity suffixes

    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
    orial
    -0.10
    ilet
    -0.09
    ãģıãĤĮ
    -0.09
     Canter
    -0.08
     noqa
    -0.08
    çĶŁäº§æĢ»åĢ¼
    -0.08
    æĦŁäºº
    -0.08
    æ±
    -0.08
    pute
    -0.08
    auge
    -0.08
    POSITIVE LOGITS
    fully
    0.13
    ful
    0.10
     ÑĢож
    0.09
    å¸ļ
    0.09
    ively
    0.09
    lessly
    0.09
    ycl
    0.09
    ãģĹãģ¾
    0.09
    è±Ĩ
    0.08
    orie
    0.08
    Activations Density 0.002%

    No Known Activations