© 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. 53146
    Prev
    Next
    INDEX
    Explanations

    hang out or hang up

    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
    ted
    -0.11
    ially
    -0.11
    ized
    -0.10
     ë¶Ī구íķĺê³ł
    -0.10
    ãģªãĤĭ
    -0.09
    éĻį温
    -0.09
    izin
    -0.09
    ected
    -0.09
    igner
    -0.09
    èݽ
    -0.09
    POSITIVE LOGITS
    xiety
    0.12
    ings
    0.10
    ä¸ŃåĽ½å¸Ĥåľº
    0.10
    alore
    0.10
    manship
    0.09
    appa
    0.09
    ulum
    0.09
    å¯ŀ
    0.09
    peed
    0.09
    abouts
    0.09
    Activations Density 0.009%

    No Known Activations