© 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. 22452
    Prev
    Next
    INDEX
    Explanations

    numbers or percentages

    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
    è¿Ļ群
    -0.10
    elig
    -0.09
    å½ĵåľ°çļĦ
    -0.09
    æĬ¢æĬĵ
    -0.09
    éļıå¤Ħ
    -0.08
    èĢģ人家
    -0.08
     locals
    -0.08
    æīĪ
    -0.08
    éĥİ
    -0.08
    pike
    -0.08
    POSITIVE LOGITS
    ç¾İåĽ½äºº
    0.10
    è½°
    0.09
    ç»ļ
    0.09
    ä¸įåłª
    0.08
     Trophy
    0.08
    ãģĬãĤĪ
    0.08
    å±Ĥåĩºä¸įç©·
    0.08
     americ
    0.08
     nett
    0.08
     Ended
    0.08
    Activations Density 0.143%

    No Known Activations