© 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. 58727
    Prev
    Next
    INDEX
    Explanations

    it has or will

    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
    æĪij们ä¸Ģèµ·
    -0.11
    éĽªå±±
    -0.10
     themselves
    -0.10
    æĪijèĩªå·±
    -0.10
    åĻİ
    -0.09
    æĪij没æľī
    -0.08
     Fury
    -0.08
    ç¼ħ
    -0.08
    代çIJĨ
    -0.08
    _LVL
    -0.08
    POSITIVE LOGITS
    èĩªèº«
    0.12
     own
    0.11
     itself
    0.10
    _own
    0.09
    Own
    0.09
    æĹĹä¸ĭçļĦ
    0.09
     Archie
    0.09
    çļĦä¸ĢåĪĩ
    0.09
     Arnold
    0.09
    梨
    0.09
    Activations Density 0.138%

    No Known Activations