© 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. 7948
    Prev
    Next
    INDEX
    Explanations

    no, only, not, less, short

    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
    è¾ľè´Ł
    -0.11
    æŀ·
    -0.10
    å»¶è¿Ł
    -0.09
    éĥ½èĥ½å¤Ł
    -0.09
    losed
    -0.09
    åıªéľĢè¦ģ
    -0.09
    è°¨
    -0.08
    azzi
    -0.08
    诧
    -0.08
    讼
    -0.08
    POSITIVE LOGITS
    abol
    0.10
    å°ĸ
    0.09
    oriously
    0.09
    寮
    0.09
    çIJĨæĥ³
    0.09
    ç©Ĺ
    0.09
    æľ¦èĥ§
    0.09
    æĬĢå¸Ī
    0.09
    usc
    0.09
    å°±æĦıåij³çĿĢ
    0.09
    Activations Density 0.410%

    No Known Activations