© 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. 53130
    Prev
    Next
    INDEX
    Explanations

    describing properties

    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
    æĪIJåĬŁçİĩ
    -0.10
    eras
    -0.09
    èĴ
    -0.08
    çĶ¨åĬĽ
    -0.08
     Instances
    -0.08
     ÑģлÑĥÑĩа
    -0.08
    æİĴéĻ¤
    -0.08
    羣çĪ±
    -0.08
     Deb
    -0.07
    èĥ¸
    -0.07
    POSITIVE LOGITS
     properties
    0.26
    çī¹æĢ§
    0.23
     Properties
    0.20
    _properties
    0.20
    properties
    0.20
     Ñģвой
    0.20
    	properties
    0.20
    (properties
    0.19
     characteristics
    0.18
    æĢ§è´¨
    0.18
    Activations Density 0.515%

    No Known Activations