© 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. 55428
    Prev
    Next
    INDEX
    Explanations

    admitting guilt or wrongdoing

    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
     Nim
    -0.09
    éĥ¤
    -0.09
    noch
    -0.09
    itm
    -0.09
     newX
    -0.08
    åĽŀå½Ĵ
    -0.08
     GAR
    -0.08
    _regression
    -0.08
     Tall
    -0.08
     Fort
    -0.08
    POSITIVE LOGITS
    æľīä»»ä½ķ
    0.11
     пÑĢизна
    0.10
    æľīä»·åĢ¼
    0.10
    åŃĺåľ¨éĹ®é¢ĺ
    0.10
    æī¿è®¤
    0.09
     acknowledge
    0.09
    çļĩåĨł
    0.09
    æĥ¶
    0.09
    çļĦåīįæıIJä¸ĭ
    0.09
    åħļçļĦé¢Ĩ导
    0.09
    Activations Density 0.047%

    No Known Activations