© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9530
    Prev
    Next
    INDEX
    Explanations

    faint

    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
    é«®
    -0.10
    çī¹æ®ĬæĥħåĨµ
    -0.09
    ç®´
    -0.09
    anson
    -0.09
    vig
    -0.08
    满æĢĢ
    -0.08
     samt
    -0.08
    urg
    -0.08
    æĦıå¤ĸ
    -0.08
    严峻
    -0.08
    POSITIVE LOGITS
    est
    0.14
    -hearted
    0.13
    è®°å¾Ĺ
    0.10
    çľĩ
    0.09
     dÃłng
    0.09
    åĩł
    0.09
     kiến
    0.09
    ly
    0.09
    è¢ħ
    0.09
     wis
    0.09
    Activations Density 0.004%

    No Known Activations