© 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. 40196
    Prev
    Next
    INDEX
    Explanations

    sliced or slices of

    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
    çĪª
    -0.11
    /INFO
    -0.10
    .DataTable
    -0.10
    ç¼ļ
    -0.10
    åĭĥåĭĥ
    -0.09
    hold
    -0.09
    èĤ¡æľ¬
    -0.09
    躯
    -0.09
    é¹°
    -0.09
    èѦåĬ¡
    -0.09
    POSITIVE LOGITS
    æű
    0.10
    kest
    0.09
    let
    0.08
    dana
    0.08
    apore
    0.08
    aped
    0.08
    åŃIJéĩĮ
    0.08
    åĸĤ
    0.08
    ron
    0.08
    arResult
    0.08
    Activations Density 0.006%

    No Known Activations