© 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. 48601
    Prev
    Next
    INDEX
    Explanations

    however followed by contrast or exception

    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
    åīģ
    -0.10
     zwar
    -0.09
    tors
    -0.09
    åĨ·æ¼ł
    -0.09
    igers
    -0.09
    orem
    -0.09
    éĤ£éº¼
    -0.08
    åĻ¤
    -0.08
    ä¸īä½į
    -0.08
    éĶ·
    -0.08
    POSITIVE LOGITS
    ãģ£ãģ¨
    0.09
     intermediary
    0.09
    å®ĥæĺ¯
    0.09
    æī§æĶ¿
    0.08
    rap
    0.08
    panse
    0.08
    еÐ
    0.08
    าม
    0.08
    umber
    0.08
    è¿ĺåľ¨
    0.08
    Activations Density 0.048%

    No Known Activations