© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37638
    Prev
    Next
    INDEX
    Explanations

    first [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
    ç®IJ
    -0.11
    é¹ĥ
    -0.11
    ãĤīãģĽ
    -0.10
    念
    -0.09
    æ±°
    -0.09
    为ä¹ĭ
    -0.09
    enido
    -0.09
    åıijæĶ¹
    -0.09
    æ¡ĤèĬ±
    -0.09
    é«ĺå°Ķ
    -0.08
    POSITIVE LOGITS
    king
    0.09
    isha
    0.09
     multit
    0.09
    æĥħåĨµä¸ĭ
    0.08
    oha
    0.08
    ONT
    0.08
    第äºĮ个
    0.08
    unker
    0.08
    -ever
    0.08
    odia
    0.08
    Activations Density 0.030%

    No Known Activations