© 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. 31038
    Prev
    Next
    INDEX
    Explanations

    classified into categories or groups

    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
    移åĬ¨ç«¯
    -0.09
    éĶ¦
    -0.09
    å°Ĭ
    -0.09
     Lord
    -0.09
    CHANT
    -0.08
    涨åģľ
    -0.08
    opsis
    -0.08
     Loy
    -0.08
    æŃ»èĢħ
    -0.08
    é¹Ĭ
    -0.08
    POSITIVE LOGITS
    ç±»åĪ«
    0.14
     category
    0.12
     categoryName
    0.10
    -category
    0.10
    çŃī级
    0.09
    Category
    0.09
    çĴİ
    0.09
    .category
    0.09
    category
    0.09
     CATEGORY
    0.09
    Activations Density 0.118%

    No Known Activations