© 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. 56667
    Prev
    Next
    INDEX
    Explanations

    in search of

    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
    åħ¬å®īéĥ¨
    -0.09
    ä¸Ĭ次
    -0.09
     Lorem
    -0.09
    æ·
    -0.08
    åĩĮ
    -0.08
     każdej
    -0.08
    éĿ¡
    -0.08
     labore
    -0.08
    ä¾Ľå¤§å®¶
    -0.08
    ä¼°åĢ¼
    -0.08
    POSITIVE LOGITS
    éĶĦ
    0.10
    é£İåIJ¹
    0.09
    赶æĿ¥
    0.09
    åij¼
    0.09
    ANI
    0.09
    æ¸ħæ·¡
    0.08
    讲究
    0.08
    ÐĪ
    0.08
    æĪĸå°Ĩ
    0.08
    _PRI
    0.08
    Activations Density 0.172%

    No Known Activations