© 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. 18943
    Prev
    Next
    INDEX
    Explanations

    the following conditions are

    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
    工夫
    -0.09
    FragmentManager
    -0.09
    æľī许å¤ļ
    -0.09
    developers
    -0.08
    bbox
    -0.08
    ãģĹãģŁãĤĤãģ®
    -0.08
    uable
    -0.08
     velvet
    -0.08
     facult
    -0.08
    EAR
    -0.08
    POSITIVE LOGITS
    çĻ½æ²Ļ
    0.10
    åĪĨåĪ«æĺ¯
    0.09
    :↵
    0.09
    :
    0.09
    æīįæĺ¯çľŁæŃ£
    0.08
    PL
    0.08
    ãģĦãģļãĤĮ
    0.08
    ç»·
    0.08
     kob
    0.08
    æĸĻ
    0.08
    Activations Density 0.184%

    No Known Activations