© 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. 30986
    Prev
    Next
    INDEX
    Explanations

    names and suffixes

    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
    æķŀ
    -0.11
    éŀħ
    -0.10
    lest
    -0.10
    èľļ
    -0.09
    strap
    -0.09
    涧
    -0.09
    硬åĮĸ
    -0.09
    æĹłç¼Ŀ
    -0.08
    å¤Ļ
    -0.08
    æ¦ľ
    -0.08
    POSITIVE LOGITS
    ç¾
    0.10
    泡泡
    0.09
    å·¢
    0.09
    olon
    0.08
    cona
    0.08
    çĶij
    0.08
    ä¼ļè§ģ
    0.08
    çĹ´
    0.08
    æļĤæŶ
    0.08
    èĽĻ
    0.08
    Activations Density 0.079%

    No Known Activations