© 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. 35664
    Prev
    Next
    INDEX
    Explanations

    copyright license

    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
    mant
    -0.11
     تصÙĨÙĬÙģ
    -0.10
    ymb
    -0.10
    åºĶçĶ¨åľºæĻ¯
    -0.10
    _CD
    -0.09
    anker
    -0.09
    dma
    -0.09
    é¡·
    -0.08
    imdi
    -0.08
    matchCondition
    -0.08
    POSITIVE LOGITS
    çķħ
    0.09
    èĥ½åĬĽå¼º
    0.09
    åį®
    0.08
    åĵĨ
    0.08
    æĢ§å¼º
    0.08
     TAR
    0.08
    çľĭä¸įè§ģ
    0.08
     Hed
    0.08
    linked
    0.08
     prefix
    0.07
    Activations Density 0.053%

    No Known Activations