© 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. 3971
    Prev
    Next
    INDEX
    Explanations

    Downing, Downs, Down

    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
    楯
    -0.10
    omic
    -0.09
     Shack
    -0.09
    æľĹ诵
    -0.09
     ÄijÆ°Æ¡ng
    -0.09
    POCH
    -0.09
    gems
    -0.09
     unsur
    -0.09
    è¿Ļè¾ĪåŃIJ
    -0.09
    èķ»
    -0.09
    POSITIVE LOGITS
    -thirds
    0.10
    ject
    0.09
    -ended
    0.09
    åħ«åŃĹ
    0.08
    ners
    0.08
    ðŁı»
    0.08
    ersist
    0.08
    enheim
    0.08
     propri
    0.08
    à¹Ĩ
    0.08
    Activations Density 0.011%

    No Known Activations