© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 19922
    Prev
    Next
    INDEX
    Explanations

    long story short

    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
    roc
    -0.09
    çļĦç§ĺå¯Ĩ
    -0.08
    çĿ£
    -0.08
    åħľ
    -0.08
    ipsis
    -0.08
    ä¿Ŀå¯Ĩ
    -0.08
    æĺ¯éĿŀ常
    -0.08
    ROC
    -0.08
    ç¿ħ
    -0.08
     jun
    -0.08
    POSITIVE LOGITS
    æ³°
    0.12
    åĮ£
    0.11
    ä¸įçĦ¶
    0.11
    æłī
    0.10
     Mim
    0.10
    çİ°åĩº
    0.09
    reh
    0.09
    æĹĹä¸ĭ
    0.09
    éħįå¥Ĺ
    0.09
    æĿĵ
    0.09
    Activations Density 0.011%

    No Known Activations