© 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. 49268
    Prev
    Next
    INDEX
    Explanations

    choose to + verb

    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
    orig
    -0.09
    å¸ĮæľĽèĥ½
    -0.09
    æĢķ
    -0.09
    (TIM
    -0.09
    è®°èĢħ
    -0.09
    æķ¢
    -0.08
    æĥ³çŁ¥éģĵ
    -0.08
     skÅĤad
    -0.08
    åıijçİ°èĩªå·±
    -0.08
    æĥ³æī¾
    -0.08
    POSITIVE LOGITS
    éĵ¤
    0.12
    izard
    0.10
     ignor
    0.09
    éĿ
    0.09
    åIJĽ
    0.09
    è¿Ļä¹Īåģļ
    0.09
    è¿Ļæł·åģļ
    0.08
    èĩ
    0.08
     Ital
    0.08
    为ä¹ĭ
    0.08
    Activations Density 0.023%

    No Known Activations