© 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. 10626
    Prev
    Next
    INDEX
    Explanations

    hard/difficult + [noun]

    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
    ping
    -0.10
    楦
    -0.09
    仪å¼ı
    -0.09
    å®īåħ¨éļIJæĤ£
    -0.09
    arend
    -0.08
    éĽ¾
    -0.08
    à¹Į
    -0.08
     Ping
    -0.08
     curtain
    -0.08
     Fen
    -0.08
    POSITIVE LOGITS
    åī²
    0.10
     доб
    0.10
    å¾Ħ
    0.10
    硬
    0.09
    åĨ³å®ļ
    0.09
    æĹ¶ä»£ä¸ŃåĽ½
    0.09
    å¾ij
    0.09
    決å®ļ
    0.09
    won
    0.09
    æĪªæŃ¢
    0.09
    Activations Density 0.013%

    No Known Activations