© 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. 58840
    Prev
    Next
    INDEX
    Explanations

    and for reasons

    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
    ermen
    -0.10
    çIJĨæĥ³çļĦ
    -0.09
    _AX
    -0.09
    ONES
    -0.08
    çīĮåŃIJ
    -0.08
     ×Ļצ×IJ
    -0.08
    oku
    -0.08
    çīĴ
    -0.08
    èĥ´
    -0.08
     Entry
    -0.08
    POSITIVE LOGITS
     reason
    0.16
    æĺ¯åĽłä¸º
    0.16
    reason
    0.15
    åİŁåĽł
    0.14
     ìĿ´ìľł
    0.14
    å°±æĺ¯åĽłä¸º
    0.14
    çļĦåİŁåĽł
    0.13
    çIJĨçĶ±
    0.13
     reasons
    0.13
    éģĵçIJĨ
    0.12
    Activations Density 0.041%

    No Known Activations