© 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. 8390
    Prev
    Next
    INDEX
    Explanations

    closing things or actions

    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
    æ·®
    -0.09
    ãĥ´ãĤ£
    -0.09
    acement
    -0.08
    ç¢Į
    -0.08
    prung
    -0.08
    çĺ¤
    -0.08
    奴
    -0.08
     görüntü
    -0.08
    IGNORE
    -0.08
    çĤ¹å¤´
    -0.08
    POSITIVE LOGITS
    åħ³éĹŃ
    0.15
    éĸī
    0.15
    éĹŃ
    0.15
     closed
    0.13
     shut
    0.13
    å°ģéĹŃ
    0.13
    closed
    0.12
     closing
    0.11
     clos
    0.11
     Closed
    0.11
    Activations Density 0.094%

    No Known Activations