© 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. 61948
    Prev
    Next
    INDEX
    Explanations

    at positions

    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
     bour
    -0.09
    chein
    -0.09
    æľĢå¿«çļĦ
    -0.09
    avy
    -0.09
    æ´»è·ĥ
    -0.09
    çĥĪ
    -0.09
    æ´¥
    -0.08
    ä¸ĺ
    -0.08
    çħ
    -0.08
    æīĺ
    -0.08
    POSITIVE LOGITS
    è¯
    0.11
    è¯įè¯Ń
    0.10
    人å±ħ
    0.09
    èĥĮçĿĢ
    0.09
    éĺIJæĺİ
    0.09
    ink
    0.08
     Archbishop
    0.08
    饵
    0.08
    æįĨ
    0.08
    umu
    0.08
    Activations Density 0.020%

    No Known Activations