© 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. 8424
    Prev
    Next
    INDEX
    Explanations

    sentence identification, linguistic rules, phrases

    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
    åŀĽ
    -0.10
    åıĮèĦļ
    -0.09
    åIJĦèĩªçļĦ
    -0.09
    åĪĪ
    -0.09
    çĽĪ
    -0.09
    è¯ŃéŁ³
    -0.09
     LANGUAGE
    -0.08
    仪
    -0.08
    åIJŁ
    -0.08
    å¼Ģæľº
    -0.08
    POSITIVE LOGITS
    ASA
    0.09
     ëĬ
    0.09
    å£ij
    0.09
     NOTICE
    0.09
    转æĬĺ
    0.09
    eselect
    0.09
    ulin
    0.09
    oler
    0.08
    roud
    0.08
    ä¹±
    0.08
    Activations Density 0.073%

    No Known Activations