© 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. 6364
    Prev
    Next
    INDEX
    Explanations

    in case of

    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
    åı©
    -0.09
    çļĦå¿ĥæĥħ
    -0.09
    éĺ¶
    -0.09
    idia
    -0.09
    ä¸įçŁ¥ä¸įè§ī
    -0.09
    auce
    -0.08
    ç®
    -0.08
    çĸ£
    -0.08
    unfold
    -0.08
    éĤ£ä¸ĢåĪ»
    -0.08
    POSITIVE LOGITS
    oeff
    0.10
     Rew
    0.08
    دÙĪ
    0.08
    åİĿ
    0.08
    agonal
    0.08
    æļ¹
    0.08
     Cha
    0.08
    شا
    0.08
    ç¬ijèĦ¸
    0.08
    æĭ¿æĿ¥
    0.07
    Activations Density 0.006%

    No Known Activations