© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21236
    Prev
    Next
    INDEX
    Explanations

    copyright notice

    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
     uÄŁ
    -0.10
    MOTE
    -0.08
    cntl
    -0.08
    丰
    -0.08
    ÙĬÙĨا
    -0.08
    esch
    -0.08
    æľªæĿ¥çļĦ
    -0.08
     Clayton
    -0.08
    åĵĨ
    -0.08
    Invocation
    -0.08
    POSITIVE LOGITS
     noqa
    0.12
    ç«¥
    0.10
    åĨ³å¿ĥ
    0.09
    ä¸Ģ个æĸ°çļĦ
    0.08
    æĺ¯æĪijçļĦ
    0.08
     Lair
    0.08
    ä¸Ģçŀ¬éĹ´
    0.08
    稼
    0.08
    иÑĢ
    0.08
    èĩªè§ī
    0.08
    Activations Density 0.683%

    No Known Activations