© 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. 9861
    Prev
    Next
    INDEX
    Explanations

    bound or reserved

    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
    éĺ¶
    -0.10
    oot
    -0.10
    éķ¿è¾Ī
    -0.09
    èĮ«
    -0.09
    æī¿æĭħ
    -0.09
    çĸ¯
    -0.09
     postings
    -0.09
    jÄĻ
    -0.09
    å¾Ĵ
    -0.08
    æľµ
    -0.08
    POSITIVE LOGITS
    ackle
    0.10
    _unlock
    0.09
    éĩįçĤ¹é¢ĨåŁŁ
    0.09
    æ¿ī
    0.09
    гÑĥ
    0.09
    itur
    0.08
     âĵĺ
    0.08
    æݢ访
    0.08
    äºĨ解ä¸Ģä¸ĭ
    0.08
    宾
    0.08
    Activations Density 0.001%

    No Known Activations