© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 60618
    Prev
    Next
    INDEX
    Explanations

    positive

    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
    Negative Logits
    iness
    -0.10
    ä¹İ
    -0.10
    spaces
    -0.10
    aye
    -0.09
    opot
    -0.09
    ÑĢ
    -0.08
    é«ĺ度éĩįè§Ĩ
    -0.08
    ittle
    -0.08
    /is
    -0.08
    otomy
    -0.08
    POSITIVE LOGITS
    -minded
    0.13
    -thinking
    0.12
     minded
    0.11
    稳妥
    0.11
    ity
    0.11
    ITIVE
    0.10
    ITY
    0.10
    -feedback
    0.10
     definite
    0.10
    idad
    0.10
    Activations Density 0.061%

    No Known Activations