© 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. 3262
    Prev
    Next
    INDEX
    Explanations

    entities and their owners

    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
    æµģæ°´
    -0.12
    æı´
    -0.11
    çī¹åĪ«æĺ¯åľ¨
    -0.10
    æİĤ
    -0.09
    ç¢ī
    -0.09
    issy
    -0.09
    åľ¨ä¸Ĭæµ·
    -0.09
    侮辱
    -0.09
    issing
    -0.09
    æīĵæĸŃ
    -0.09
    POSITIVE LOGITS
    usch
    0.10
    对åħ¶è¿Ľè¡Į
    0.09
    åįķä¸Ģ
    0.09
     owning
    0.09
    对该
    0.09
    åħ¬åı¸å°Ĩ
    0.09
     corporation
    0.09
     owns
    0.08
    çĪ·
    0.08
    对åħ¶
    0.08
    Activations Density 0.124%

    No Known Activations