© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52402
    Prev
    Next
    INDEX
    Explanations

    people and relationships

    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
    æĢ§å¼º
    -0.10
    寿
    -0.09
    olon
    -0.09
    (Properties
    -0.09
    æĻ®éģį
    -0.08
    ç»ıåħ¸çļĦ
    -0.08
    DET
    -0.08
    زاÙĨ
    -0.08
    ificate
    -0.08
    esimal
    -0.08
    POSITIVE LOGITS
     лÑİбим
    0.13
     ëį
    0.11
    å±ŀäºİèĩªå·±
    0.11
     own
    0.10
    èĩªå·±çļĦ
    0.10
    çī¹éķ¿
    0.10
    èĽĭ
    0.10
    æĬĬèĩªå·±çļĦ
    0.09
    éºĴ
    0.09
    äºĨèĩªå·±çļĦ
    0.09
    Activations Density 0.330%

    No Known Activations