© 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. 59454
    Prev
    Next
    INDEX
    Explanations

    binning data, files, or code

    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
    ech
    -0.13
    ening
    -0.11
    ened
    -0.11
    anos
    -0.10
    Thrown
    -0.09
    ãģ¨ãĤĬ
    -0.09
    ably
    -0.09
    ãģ¶ãĤĬ
    -0.09
    еÑĩен
    -0.09
    è£Ķ
    -0.09
    POSITIVE LOGITS
    ocular
    0.16
    åŃIJéĩĮ
    0.14
     Laden
    0.13
    omial
    0.13
    itial
    0.11
    aries
    0.10
    ning
    0.10
    å¿ĥåºķ
    0.09
    á»iji
    0.09
    èĩ¾
    0.09
    Activations Density 0.034%

    No Known Activations