© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 30966
    Prev
    Next
    INDEX
    Explanations

    Superlatives and ordinals

    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
    æĩĬ
    -0.10
    .dw
    -0.10
    æı©
    -0.10
    å½¹
    -0.09
    ény
    -0.09
    ä¸Ńåįİ
    -0.09
    çĿ£åĬŀ
    -0.09
    æĬĺæī£
    -0.09
    SMART
    -0.09
    าย
    -0.09
    POSITIVE LOGITS
    æ·¹
    0.09
     cá
    0.08
    çŁ³æ¦´
    0.08
     popcorn
    0.08
    iv
    0.08
    åİĭ
    0.08
    prs
    0.08
    .pp
    0.08
    éĢģæĿ¥
    0.08
    elen
    0.08
    Activations Density 0.077%

    No Known Activations