© 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. 36451
    Prev
    Next
    INDEX
    Explanations

    take/takes followed by prepositions or articles

    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
    entes
    -0.09
    alink
    -0.09
    isse
    -0.09
    éĵº
    -0.08
     Lords
    -0.08
    éģĩ
    -0.08
    /&
    -0.08
    èݽ
    -0.08
    åij¼åIJģ
    -0.08
    ninger
    -0.08
    POSITIVE LOGITS
    volt
    0.09
     Bellev
    0.09
     Berm
    0.09
    æ¯ħåĬĽ
    0.09
    ç´łè´¨
    0.08
    stile
    0.08
    èī¯å¿ĥ
    0.08
    åĽŀåij³
    0.08
    æ²®
    0.08
    åĹħ
    0.08
    Activations Density 0.037%

    No Known Activations