© 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. 50397
    Prev
    Next
    INDEX
    Explanations

    analyzed data

    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
    eli
    -0.09
    gg
    -0.09
    ĥģ
    -0.09
    ToFile
    -0.09
    etto
    -0.09
    ferences
    -0.08
    çĩ®
    -0.08
    ä»Ĺ
    -0.08
    INCLUDE
    -0.08
    лоÑĤ
    -0.08
    POSITIVE LOGITS
     existing
    0.10
    çİ°æľīçļĦ
    0.10
    çİ°æľī
    0.10
    表çĻ½
    0.10
    æĺ¯åIJ¦åŃĺåľ¨
    0.09
    æĺ¯åIJ¦æľī
    0.09
    åij¨åĽ´çļĦ
    0.09
    å®ŀéĻħæĥħåĨµ
    0.09
    åģļå·¥
    0.09
    è¿Ļ个éĹ®é¢ĺ
    0.09
    Activations Density 0.176%

    No Known Activations