© 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. 20381
    Prev
    Next
    INDEX
    Explanations

    primary sources

    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
    duk
    -0.09
    çľ©
    -0.08
    _PB
    -0.08
    é¹ij
    -0.08
    STAT
    -0.08
    绪
    -0.07
    绵
    -0.07
    éĹ»
    -0.07
    亮çĤ¹
    -0.07
     STAT
    -0.07
    POSITIVE LOGITS
    åı²æĸĻ
    0.15
    åı¯ä¿¡
    0.11
    èµĦæĸĻ
    0.11
    à¹ģหล
    0.10
    åİŁä»¶
    0.10
    æĸĩçī©
    0.10
    æĿIJæĸĻ
    0.09
    ADIO
    0.09
    å¾Īå¼Ģå¿ĥ
    0.09
    åĩºåľŁ
    0.09
    Activations Density 0.067%

    No Known Activations