© 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. 17609
    Prev
    Next
    INDEX
    Explanations

    first-person "i" followed by verbs

    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
    æģŃ
    -0.11
    eing
    -0.10
    æĦ£
    -0.10
    妻åŃIJ
    -0.09
    éĶ´
    -0.09
    èĢģ头
    -0.09
    é¬ĵ
    -0.09
     ss
    -0.09
     thơ
    -0.09
    èĢģå©Ĩ
    -0.09
    POSITIVE LOGITS
    nia
    0.08
    滥
    0.08
    iae
    0.08
    说æľį
    0.08
     rede
    0.08
    æľĢåĸľæ¬¢
    0.08
    è¶ħé¢Ŀ
    0.08
     Meyer
    0.08
    æĽ´å¤ļ
    0.08
    åī
    0.08
    Activations Density 0.044%

    No Known Activations