© 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. 26599
    Prev
    Next
    INDEX
    Explanations

    arrives at, heard a, attacks targeting

    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
    æľªå¿ħ
    -0.09
    æ¹ĸåĮº
    -0.09
    ISO
    -0.09
    è¸ŀ
    -0.08
    æľĢ容æĺĵ
    -0.08
    ä¸ĩä¸Ģ
    -0.08
    5
    -0.08
    3
    -0.08
    æ¯Ļ
    -0.08
    è¿Ħä»Ĭ
    -0.08
    POSITIVE LOGITS
    _Internal
    0.10
    缤
    0.10
    rü
    0.09
    诬
    0.09
    oice
    0.09
     Flux
    0.09
    说éģĵ
    0.09
    æĿ¥åĪ°äºĨ
    0.09
     모ìĬµ
    0.08
    INTERNAL
    0.08
    Activations Density 0.040%

    No Known Activations