© 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. 61583
    Prev
    Next
    INDEX
    Explanations

    Naugatuck Valley

    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
     sw
    -0.10
    èĤļåŃIJ
    -0.10
    æĥ©ç½ļ
    -0.09
     ÑģовеÑĢÑĪен
    -0.09
    меÑģ
    -0.09
    æݾ
    -0.09
    esan
    -0.09
    æĥ©
    -0.08
    ç¬ķ
    -0.08
    .TableName
    -0.08
    POSITIVE LOGITS
    éĴ¢éĵģ
    0.12
    åĨ¶éĩij
    0.12
     steel
    0.12
     metals
    0.12
    åĨ¶çĤ¼
    0.12
     Steel
    0.12
    éĴ¢
    0.11
    çŁ¿
    0.11
    éĴ¢åİĤ
    0.11
    éĴ¢æĿIJ
    0.11
    Activations Density 0.164%

    No Known Activations