© 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. 51446
    Prev
    Next
    INDEX
    Explanations

    extortion, blackmail, demands, coerced

    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
    éĢIJ
    -0.12
    çĨ¬
    -0.09
    ç¿°
    -0.09
     Lotus
    -0.08
     Cutter
    -0.08
    åłĻ
    -0.08
    çģ¸
    -0.08
    abo
    -0.08
    æĿ¥å®¾
    -0.08
    åĪ°çİ°åľ¨
    -0.08
    POSITIVE LOGITS
    微微
    0.10
     extortion
    0.09
    press
    0.09
     demands
    0.09
    亲çĪ±
    0.08
    ÐĤ
    0.08
    çłĿ
    0.08
     blackmail
    0.08
    约è°Ī
    0.08
    æħ
    0.08
    Activations Density 0.050%

    No Known Activations