© 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. 38746
    Prev
    Next
    INDEX
    Explanations

    Dragon Point, Finland

    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
    åĨħè¡£
    -0.10
    å®ĺåı¸
    -0.10
    åºŁè¯Ŀ
    -0.10
    æŀľ
    -0.09
    ç£ĭ
    -0.09
    ãĥīãĥ¬ãĤ¹
    -0.09
    ç¨į
    -0.08
    æĺ¾ç¤º
    -0.08
    ãĤ¶
    -0.08
     converse
    -0.08
    POSITIVE LOGITS
     metaphor
    0.11
    èĪ¬çļĦ
    0.10
    尾巴
    0.09
    ä¸įèµ·
    0.09
    åĸ»
    0.09
    æ´Ļ
    0.08
    ä¼¼çļĦ
    0.08
    è½Ń
    0.08
    éĻ¨
    0.08
    iculos
    0.08
    Activations Density 0.297%

    No Known Activations