© 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. 19355
    Prev
    Next
    INDEX
    Explanations

    Greek and other non-Latin scripts

    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
    ستر
    -0.08
    linger
    -0.08
     potency
    -0.08
    ائØŃ
    -0.08
    è°¨
    -0.08
    羣æŃ£
    -0.08
    èľľ
    -0.08
    odesk
    -0.08
    æ±
    -0.08
    seek
    -0.08
    POSITIVE LOGITS
    iad
    0.10
    æĿĥ
    0.10
    bable
    0.09
     ÑĢоман
    0.09
    inciple
    0.09
    ersions
    0.09
    对æİ¥
    0.08
    说äºĨ
    0.08
    erness
    0.08
    æ¯ķç«Łæĺ¯
    0.08
    Activations Density 0.048%

    No Known Activations