© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 48640
    Prev
    Next
    INDEX
    Explanations

    Taliban and mafia groups

    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
    zza
    -0.11
    ве
    -0.09
    ãģĵãģ¨ãģĮãģ§ãģį
    -0.09
    ernals
    -0.09
    mitt
    -0.09
    ders
    -0.09
    lotte
    -0.08
    .bounds
    -0.08
    plans
    -0.08
    ollen
    -0.08
    POSITIVE LOGITS
    ðŁı»
    0.11
    0.09
    /or
    0.09
    ä¸ļä½Ļ
    0.09
    è¦ģçŁ¥éģĵ
    0.08
    纵深
    0.08
    wagon
    0.08
    éĹ®è´£
    0.08
     ReSharper
    0.08
    ãģ»ãģĨ
    0.08
    Activations Density 0.001%

    No Known Activations