© 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. 50955
    Prev
    Next
    INDEX
    Explanations

    l and following letters

    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
    ä¸Ģ头
    -0.09
    osta
    -0.09
    éķı
    -0.09
    æĥ®
    -0.08
    å¿Ĵ
    -0.08
    ä¼ī
    -0.08
    å¸Ļ
    -0.08
    为æ°ij
    -0.08
     Harding
    -0.08
    æĢİ麼
    -0.08
    POSITIVE LOGITS
    人å¿ĥ
    0.11
    çĶ¨è½¦
    0.08
    fern
    0.08
    Toggle
    0.08
    ifes
    0.08
    erness
    0.08
    lo
    0.08
    olated
    0.08
    anguage
    0.08
    å¿ĥéĩĮ
    0.08
    Activations Density 0.041%

    No Known Activations