© 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. 33139
    Prev
    Next
    INDEX
    Explanations

    as far as I know

    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
    eya
    -0.11
    onna
    -0.09
    æľºç»Ħ
    -0.09
    ä»Ķ
    -0.09
    demand
    -0.09
    å¤įæ´»
    -0.09
    åĿı
    -0.09
    å²±
    -0.09
    涧
    -0.09
    象å¾ģ
    -0.09
    POSITIVE LOGITS
    æĪijçľĭ
    0.10
    çľĭæĿ¥
    0.10
    åı¯çŁ¥
    0.10
    æĪij没
    0.10
    èĢĮè¨Ģ
    0.10
    çŁ¥éģĵèĩªå·±
    0.10
    éĥ½çŁ¥éģĵ
    0.09
    è¿Ħä»Ĭ
    0.09
    æĪijçŁ¥éģĵ
    0.09
    æĥħåĨµæĿ¥çľĭ
    0.09
    Activations Density 0.071%

    No Known Activations