© 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. 10165
    Prev
    Next
    INDEX
    Explanations

    silence, secrecy, confidentiality

    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
    æĤĴ
    -0.10
    FunctionFlags
    -0.09
    主åĬŀ
    -0.09
    åĪĨå±Ģ
    -0.09
     Seat
    -0.09
    éĤł
    -0.09
    coli
    -0.09
    åIJ¸å¼ķ
    -0.09
    éłĵ
    -0.09
    /desktop
    -0.09
    POSITIVE LOGITS
    ä¿Ŀå¯Ĩ
    0.20
     silence
    0.15
    æ²īé»ĺ
    0.14
    ç¼Ħ
    0.14
     confidentiality
    0.13
     secrecy
    0.12
    sil
    0.12
    ä¸į让
    0.12
    éļIJçŀĴ
    0.12
     Confidential
    0.12
    Activations Density 0.104%

    No Known Activations