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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52109
    Prev
    Next
    INDEX
    Explanations

    give or take

    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
    åĴĮæĬĢæľ¯
    -0.09
    icate
    -0.09
    绣
    -0.09
    å¢ŀéĢŁ
    -0.09
     wastes
    -0.08
    alore
    -0.08
    æħij
    -0.08
    è¡Įäºĭ
    -0.08
    mak
    -0.08
    ÑİÑģÑĮ
    -0.08
    POSITIVE LOGITS
    ç´¢åıĸ
    0.15
    è¿Ķè¿ĺ
    0.10
    åĽŀé¦Ī
    0.10
    åĽŀæĬ¥
    0.10
    å²±
    0.09
    íĤ·
    0.09
    itti
    0.09
    äºĪ
    0.09
    .fromString
    0.09
    çļĦè§Ĵ度
    0.09
    Activations Density 0.055%

    No Known Activations