© 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. 34000
    Prev
    Next
    INDEX
    Explanations

    dissolve, offic, section

    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
    å°İ
    -0.10
    åħĪ导
    -0.09
    stro
    -0.09
    erd
    -0.09
    ä¸Ģåı·
    -0.09
    ç¿Ł
    -0.09
    cis
    -0.09
    ãĥĬãĥ³
    -0.09
    éĤĦæĺ¯
    -0.09
    LEAN
    -0.08
    POSITIVE LOGITS
    å¤ĩæĪĺ
    0.10
    解æĶ¾
    0.10
    踵
    0.09
    ickle
    0.09
    æĮ£
    0.08
     comp
    0.08
    èĢģçĪ·
    0.08
    éĩĬæĶ¾
    0.08
    èĢ³æľµ
    0.08
    _qual
    0.08
    Activations Density 0.132%

    No Known Activations