© 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. 8165
    Prev
    Next
    INDEX
    Explanations

    selecting or choosing something

    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
     anytime
    -0.10
    oft
    -0.08
    å¤Ħæĸ¹
    -0.08
    çİ°å®ŀ
    -0.08
    amos
    -0.08
    SAVE
    -0.08
    èĹı
    -0.08
     Weiner
    -0.08
    éļıæŶ
    -0.08
     parch
    -0.08
    POSITIVE LOGITS
    çļĦéĢīæĭ©
    0.16
    éĢīæĭ©
    0.15
    åIJĪéĢĤçļĦ
    0.15
    éĢĤåIJĪèĩªå·±
    0.14
    éģ¸æĵĩ
    0.13
    éĢīæĭ©äºĨ
    0.13
    åIJĪéĢĤ
    0.12
     вÑĭбиÑĢа
    0.12
     choice
    0.12
     choosing
    0.11
    Activations Density 0.111%

    No Known Activations