© 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. 28111
    Prev
    Next
    INDEX
    Explanations

    approving controversial revisions

    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
    广æĴŃ
    -0.10
    lama
    -0.10
    agg
    -0.09
    .fml
    -0.08
    è¿Ļè¾¹
    -0.08
    åİĸ
    -0.08
    åĬŀ好
    -0.08
    æ¹Ľ
    -0.08
    digest
    -0.08
    iscard
    -0.07
    POSITIVE LOGITS
    æĽ¾åľ¨
    0.11
    缺åı£
    0.10
    举
    0.10
    åĨ³çŃĸ
    0.10
    ç½ij讯
    0.10
    gt
    0.10
    èµĿ
    0.09
    èĪī
    0.09
    åĨ³è®®
    0.09
    å¥ı
    0.09
    Activations Density 0.069%

    No Known Activations