© 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. 52263
    Prev
    Next
    INDEX
    Explanations

    spatial or conceptual divisions

    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
    åįĥåı¤
    -0.09
    é«ĺæĸ°
    -0.08
    ead
    -0.08
    宥
    -0.08
    å®ĺåı¸
    -0.08
    ãi
    -0.08
    _SIG
    -0.08
    ãģ¤ãģ¤
    -0.08
    ги
    -0.08
    èĥ¶
    -0.08
    POSITIVE LOGITS
     thereof
    0.18
    让å®ĥ
    0.11
    使ä¹ĭ
    0.11
    æĬĬå®ĥ
    0.10
     ниÑħ
    0.10
    对åħ¶è¿Ľè¡Į
    0.10
    èĪĴ
    0.10
    å°Ĩåħ¶
    0.09
     него
    0.09
    使åħ¶
    0.09
    Activations Density 0.489%

    No Known Activations