© 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. 18796
    Prev
    Next
    INDEX
    Explanations

    go to or from

    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
    çıį
    -0.09
    oria
    -0.09
    APE
    -0.09
    зн
    -0.09
    åĩī
    -0.09
    ÙĬÙĩ
    -0.08
    hana
    -0.08
    dre
    -0.08
    æµģæ°´
    -0.08
    .Inst
    -0.08
    POSITIVE LOGITS
    é«ĺå±Ĥ次
    0.08
    icom
    0.08
    @qq
    0.08
    æĭ¥
    0.08
     Granted
    0.08
    åģľçķĻåľ¨
    0.08
    åĽ´çĿĢ
    0.08
    æ¼Ĥ亮çļĦ
    0.08
    æľī人
    0.08
    é«ĺ端
    0.08
    Activations Density 0.021%

    No Known Activations