© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 22915
    Prev
    Next
    INDEX
    Explanations

    of followed by specific names

    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
    èİİ
    -0.08
    еÑĨ
    -0.08
    erti
    -0.08
    å¯Ł
    -0.08
    èĥİ
    -0.08
    éļĶçĿĢ
    -0.08
    illi
    -0.08
    涵
    -0.08
    eni
    -0.07
    _tac
    -0.07
    POSITIVE LOGITS
    åĹĵåŃIJ
    0.10
    éĶ¥
    0.10
    æijĩ
    0.09
    æľīæĿ¡ä»¶çļĦ
    0.09
    ãģ°ãģĭãĤĬ
    0.09
    른
    0.09
    泫
    0.09
    æĥ³è±¡åĬĽ
    0.09
    rias
    0.09
    èĩ»
    0.09
    Activations Density 0.237%

    No Known Activations