© 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. 12304
    Prev
    Next
    INDEX
    Explanations

    where the

    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
    tri
    -0.09
    ASS
    -0.09
    sp
    -0.09
    EMS
    -0.08
    rie
    -0.08
    .TryParse
    -0.08
    以ä¸ĭç®Ģ称
    -0.08
    son
    -0.08
    iph
    -0.08
    ê¹IJ
    -0.08
    POSITIVE LOGITS
    abouts
    0.23
    upon
    0.20
    ever
    0.17
    fore
    0.15
    -ever
    0.15
    soever
    0.14
    upo
    0.12
    'er
    0.11
    eve
    0.11
    /if
    0.10
    Activations Density 0.085%

    No Known Activations