© 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. 58049
    Prev
    Next
    INDEX
    Explanations

    domain names

    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
    çĩĥ
    -0.10
    legg
    -0.10
    çľ¨çľ¼
    -0.10
    _GU
    -0.09
    éĶ·
    -0.09
    æīī
    -0.09
    æĿ³
    -0.09
    inite
    -0.08
    çļĭ
    -0.08
    slash
    -0.08
    POSITIVE LOGITS
    æĸ°éĹ»ä¸Ńå¿ĥ
    0.10
    好åIJ§
    0.10
    æĻ®éģį
    0.09
    家åºŃæķĻèĤ²
    0.08
    解å¼Ģ
    0.08
    æŃ¢æįŁ
    0.08
    åħ¬è®¤
    0.08
    éĩįå·¥
    0.08
    ornings
    0.08
    iant
    0.08
    Activations Density 0.009%

    No Known Activations