© 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. 2047
    Prev
    Next
    INDEX
    Explanations

    groups of three or five

    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
    éĶ£
    -0.09
    æµ·ä¸Ĭ
    -0.09
    èĬŁ
    -0.09
    æĭĪ
    -0.09
    è¿ĻæĿ¡è·¯
    -0.09
    æĥ³åİ»
    -0.09
    KeySpec
    -0.08
    ÑĩаÑģ
    -0.08
    èĽ¤
    -0.08
    .Keys
    -0.08
    POSITIVE LOGITS
    ä¸ī人
    0.11
    仨
    0.11
    ä¹Łä¸įä¾ĭå¤ĸ
    0.09
    两人
    0.08
    ä¸Ģä¸ĩ
    0.08
    å¡ij
    0.08
    hed
    0.08
    å¤ļ人
    0.08
    akis
    0.08
     Pascal
    0.08
    Activations Density 0.073%

    No Known Activations