© 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. 11095
    Prev
    Next
    INDEX
    Explanations

    interrogating or impersonating

    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
    äºĨèĩªå·±çļĦ
    -0.11
    å®ŀçĶ¨
    -0.10
    éļľ
    -0.09
    对ä»ĺ
    -0.09
    ç¼®
    -0.09
    resse
    -0.09
    åIJĮæľŁ
    -0.09
    èĩªå¾ĭ
    -0.09
    ä¾¿å®ľ
    -0.09
    _RG
    -0.09
    POSITIVE LOGITS
    ми
    0.09
    mer
    0.09
    奢
    0.08
    ible
    0.08
    ature
    0.08
     оÑģновном
    0.08
    usty
    0.08
     posi
    0.07
    à¸ĵ
    0.07
    mers
    0.07
    Activations Density 0.005%

    No Known Activations