© 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. 38654
    Prev
    Next
    INDEX
    Explanations

    out of

    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
    çĮĹ
    -0.11
    大èĩ£
    -0.10
    ENTS
    -0.09
    æ²ĸ
    -0.09
    ents
    -0.08
     strike
    -0.08
    æ°ĽåĽ´
    -0.08
    éĽį
    -0.08
    ording
    -0.08
    åIJĮæł·
    -0.08
    POSITIVE LOGITS
    éŨçªĹ
    0.11
    æĮĤåľ¨
    0.10
     doors
    0.10
    éķ±
    0.09
    çªĹæĪ·
    0.09
     Doors
    0.09
    大éŨ
    0.09
    ãģ£ãģ
    0.09
    éĺĨ
    0.09
    doors
    0.08
    Activations Density 0.073%

    No Known Activations