© 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. 21852
    Prev
    Next
    INDEX
    Explanations

    is/are describing features

    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
    æģ¨
    -0.11
    åħīèį£
    -0.11
    çļ±
    -0.10
    åıijæĮ¥ä½ľçĶ¨
    -0.10
    æī¯
    -0.10
    æĪĺ
    -0.09
    æĪIJåĬŁçİĩ
    -0.09
    resse
    -0.09
    çļĦåį°è±¡
    -0.09
    ç¼Ń
    -0.09
    POSITIVE LOGITS
    éħįä¸Ĭ
    0.09
    éĢĤåIJĪ
    0.09
    éĩĩçĶ¨
    0.08
    æIJŃéħį
    0.08
    éħį
    0.08
    éĩĩçĶ¨äºĨ
    0.07
     Dominion
    0.07
    ale
    0.07
    åĶ¯ä¸ĢçļĦ
    0.07
     depos
    0.07
    Activations Density 0.061%

    No Known Activations