© 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. 33715
    Prev
    Next
    INDEX
    Explanations

    names, authors, or suffixes

    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
    æĬĸ
    -0.11
    èµ³
    -0.11
    好çİ©
    -0.10
    èĵģ
    -0.10
    jay
    -0.10
    æ¤ħ
    -0.10
    ç¥Ł
    -0.09
    -ie
    -0.09
    ä¸į容
    -0.09
    æĶ¾å¼Ģ
    -0.09
    POSITIVE LOGITS
    ocs
    0.10
    夫人
    0.10
    è¾Ľ
    0.09
    åħ¶ä½Ļ
    0.09
    enberg
    0.08
    ingles
    0.08
    occ
    0.08
     Chron
    0.08
    礼
    0.08
     flight
    0.08
    Activations Density 0.007%

    No Known Activations