© 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. 31246
    Prev
    Next
    INDEX
    Explanations

    reasons why

    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
    ç½ķ
    -0.10
    èĻ¬
    -0.09
    ç»Ħç»ĩé¢Ĩ导
    -0.09
     Bail
    -0.09
    æĹ¥äº§
    -0.09
    ارد
    -0.08
    æŶåĪ»
    -0.08
     somehow
    -0.08
    å°¼åħĭ
    -0.08
    oti
    -0.08
    POSITIVE LOGITS
     reasons
    0.12
     Reasons
    0.11
     Gründe
    0.10
    æľīçļĦ人
    0.10
    ä¼ļåĩºçİ°
    0.09
    ä¸įçĶ±
    0.09
    ä»Ģä¹ĪåİŁåĽł
    0.09
    oser
    0.09
    orama
    0.09
     Mans
    0.09
    Activations Density 0.062%

    No Known Activations