© 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. 47725
    Prev
    Next
    INDEX
    Explanations

    quantities like 'or so'

    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
     Falcon
    -0.10
    æ·ħ
    -0.10
    è·ĥ
    -0.09
    å¿«
    -0.09
    .Skip
    -0.09
    GES
    -0.09
    ÅŁÄ±
    -0.09
    _eg
    -0.09
    å¼ķèµ·
    -0.08
    å®ŀæĸ½æĸ¹æ¡Ī
    -0.08
    POSITIVE LOGITS
    ä¹Į
    0.10
    çŁ£
    0.09
    maal
    0.09
    abouts
    0.08
    ãģIJãĤīãģĦ
    0.08
    çļĦéĩį大
    0.08
    椴
    0.08
     alike
    0.08
    éļ°
    0.08
    æľªç»ı
    0.07
    Activations Density 0.112%

    No Known Activations