© 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. 3165
    Prev
    Next
    INDEX
    Explanations

    former East Germany

    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
    æīĪ
    -0.09
    æ¸İ
    -0.09
    åĿij
    -0.09
    çľģæķĻèĤ²
    -0.09
    壮
    -0.09
    ç¥ŀè¯Ŀ
    -0.08
    ä¾ł
    -0.08
    åĩ¿
    -0.08
    æł¡
    -0.08
    atu
    -0.08
    POSITIVE LOGITS
    缣
    0.11
    èĭıèģĶ
    0.10
    碱
    0.10
     alkal
    0.09
     Cold
    0.09
    ç¼ĿéļĻ
    0.09
    allee
    0.09
    å¢Ļ
    0.09
    æĭ±
    0.08
    竦
    0.08
    Activations Density 0.138%

    No Known Activations