© 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. 31343
    Prev
    Next
    INDEX
    Explanations

    return to

    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
    å³Ļ
    -0.09
    æŃ£å®Ĺ
    -0.09
    isi
    -0.09
    gage
    -0.08
    gone
    -0.08
     ESC
    -0.08
    gin
    -0.08
    ERA
    -0.08
    çªģå¦Ĥåħ¶
    -0.08
     PÃ¥
    -0.08
    POSITIVE LOGITS
    ance
    0.11
    ista
    0.11
    ä¸įäºĨ
    0.10
    ä¸Ģè¶Ł
    0.10
     uwagÄĻ
    0.09
    OrCreate
    0.09
    istas
    0.09
    OrUpdate
    0.09
    à¹īà¸Ńย
    0.09
    erer
    0.09
    Activations Density 0.052%

    No Known Activations