© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8895
    Prev
    Next
    INDEX
    Explanations

    defense system

    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
    æĪ¿
    -0.09
    ãģ¨ãģĦãģĨãģ®ãģ¯
    -0.09
    ä¸į说
    -0.08
    walls
    -0.08
    åĨįä¸Ģ次
    -0.08
    çļĦæīĭ
    -0.08
    æĭĶ
    -0.08
    å¹³åĿ¦
    -0.08
    çħ¤
    -0.08
    .Content
    -0.08
    POSITIVE LOGITS
    饵
    0.11
    èĭĩ
    0.10
    åĬ¨ä½ľ
    0.09
    æľīæķĪæĢ§
    0.09
    (actions
    0.09
    çĵ£
    0.09
    èĢĮå·²
    0.09
    æ¼Ĥæµģ
    0.09
    åĻ¨åħ·
    0.09
    COPE
    0.08
    Activations Density 0.096%

    No Known Activations