© 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. 36573
    Prev
    Next
    INDEX
    Explanations

    code and version numbers

    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
    engo
    -0.10
    çĶ³
    -0.10
    ä¸ŃåİŁ
    -0.08
    è¦ģ说
    -0.08
    AYER
    -0.08
    å«ĸ
    -0.08
    ento
    -0.08
    åĶı
    -0.08
     Herr
    -0.08
    contra
    -0.08
    POSITIVE LOGITS
    åĭ¤åĬ³
    0.09
    bare
    0.09
    asks
    0.08
    Ñij
    0.08
    edly
    0.08
    æī¾äºĨ
    0.08
    éķ¿å¾Ĺ
    0.08
    preh
    0.08
    heet
    0.08
    ache
    0.08
    Activations Density 0.018%

    No Known Activations