© 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. 2696
    Prev
    Next
    INDEX
    Explanations

    word endings forming 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
    å¯IJ
    -0.10
    ÑĥÑģÑĤ
    -0.08
    åĨ¬åŃ£
    -0.08
    ceptive
    -0.08
    åĨĮ
    -0.08
    etri
    -0.08
    åĨ¬å¤©
    -0.08
    лиÑĨ
    -0.08
    æŃ¥æŃ¥
    -0.08
    äºİæĺ¯
    -0.08
    POSITIVE LOGITS
    worth
    0.11
    ham
    0.10
    ley
    0.10
    bury
    0.10
    ington
    0.10
    ies
    0.09
    by
    0.09
    comb
    0.09
    field
    0.09
    ough
    0.09
    Activations Density 0.327%

    No Known Activations