© 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. 3491
    Prev
    Next
    INDEX
    Explanations

    but, rather, instead

    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
    à¹ģà¸Ľ
    -0.08
    å¦Ĥæŀľè¯´
    -0.08
    æĽ¿
    -0.08
    ç²Ĵ
    -0.08
    èıĬèĬ±
    -0.08
    å¸ĪçĪ¶
    -0.08
     Universe
    -0.08
    uos
    -0.08
    亲çĪ±
    -0.08
    缸æ¯Ķä¹ĭä¸ĭ
    -0.08
    POSITIVE LOGITS
    èĢĮæĺ¯
    0.23
     sondern
    0.17
     sino
    0.12
    ä½Ĩåį´
    0.12
    ä½Ĩä¹Ł
    0.11
    ä½Ĩä»ĸ
    0.11
    ä½Ĩæĺ¯å¯¹äºİ
    0.11
    ä½Ĩæĺ¯ä»ĸ
    0.10
     but
    0.10
    ä½Ĩæĺ¯æĪij
    0.09
    Activations Density 0.049%

    No Known Activations