© 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. 56503
    Prev
    Next
    INDEX
    Explanations

    probability and speculation

    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
    ä¿Ĺ称
    -0.09
    ä¿Ĺ
    -0.09
     Part
    -0.08
    iaux
    -0.08
    ander
    -0.08
    éĻ¡
    -0.08
    为åķ¥
    -0.08
    干线
    -0.08
    è¦ģä¹Ī
    -0.08
    pressor
    -0.07
    POSITIVE LOGITS
    .getLine
    0.10
    没éĹ®é¢ĺ
    0.10
    æĸ¯åŁº
    0.09
    åŁºåĩĨ
    0.09
    ä¼ļéķ¿
    0.09
    éĤĦæľī
    0.09
    ForObject
    0.09
    ä½łä¼ļ
    0.09
    à¥į
    0.09
    hood
    0.09
    Activations Density 0.041%

    No Known Activations