© 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. 62670
    Prev
    Next
    INDEX
    Explanations

    two or two times

    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
    è°
    -0.09
    çħİ
    -0.09
    çģ«èĬ±
    -0.09
    ä¸Ńåħ³
    -0.09
    驾
    -0.09
    é¢ł
    -0.08
     fals
    -0.08
    è¿Ļä¹Īä¹ħ
    -0.08
    å·²ç»ıè¶ħè¿ĩ
    -0.08
     Harley
    -0.08
    POSITIVE LOGITS
    ä¸Ģåī¯
    0.11
    (piece
    0.10
    äºĽ
    0.09
    ä¸įçŁ¥
    0.09
    è¿Ļæł·çļĦ
    0.09
    ä¸Ģç»Ħ
    0.08
    ãģĭãĤĤ
    0.08
    æĪĸå¤ļ
    0.08
     pieces
    0.08
    个
    0.08
    Activations Density 0.084%

    No Known Activations