© 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. 30488
    Prev
    Next
    INDEX
    Explanations

    source

    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
     Jennings
    -0.10
    çĪ¶åŃIJ
    -0.10
     ÑĢей
    -0.09
    éķĽ
    -0.08
    edi
    -0.08
     Prem
    -0.08
    å¥¥åľ°
    -0.08
    éĤ¨
    -0.08
    ãģĮåĩº
    -0.08
    _manifest
    -0.08
    POSITIVE LOGITS
     sources
    0.29
    æĿ¥æºIJ
    0.25
     иÑģÑĤоÑĩник
    0.25
     Sources
    0.25
     nguá»ĵn
    0.25
    sources
    0.24
     source
    0.24
     иÑģÑĤоÑĩ
    0.23
    Sources
    0.22
    source
    0.22
    Activations Density 0.227%

    No Known Activations