© 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. 7001
    Prev
    Next
    INDEX
    Explanations

    "tell you" or "say"

    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
    ç²
    -0.09
    çľĭåĪ°äºĨ
    -0.09
    大ä¼Ĺ
    -0.09
    çļĦçľ¼
    -0.09
    å¿«
    -0.09
    娩
    -0.09
    ТÐŀ
    -0.08
    å·±
    -0.08
     Ùĥب
    -0.08
    .Pool
    -0.08
    POSITIVE LOGITS
    åijĬè¯ī大家
    0.10
    åijĬè¯īä½ł
    0.10
     Truth
    0.09
    arna
    0.09
    ç»ĻæĤ¨
    0.09
     truth
    0.09
    chron
    0.09
    ç»Ļä½łä»¬
    0.09
    çĦ°
    0.09
    Truth
    0.09
    Activations Density 0.041%

    No Known Activations