© 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. 31023
    Prev
    Next
    INDEX
    Explanations

    detecting presence or changes

    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
    лиÑĤ
    -0.11
    æľ¬èº«çļĦ
    -0.09
    maal
    -0.09
    æľ¬èº«å°±
    -0.08
    سب
    -0.08
    isphere
    -0.08
    TIM
    -0.08
    çİ«
    -0.08
    娩
    -0.08
    äºī
    -0.08
    POSITIVE LOGITS
    éļIJèĹı
    0.15
    çļĦåŃĺåľ¨
    0.13
     hidden
    0.13
     налиÑĩ
    0.12
    迹象
    0.11
    æĺ¯åIJ¦åŃĺåľ¨
    0.11
     presence
    0.11
    hidden
    0.11
    ç»Ĩå¾®
    0.11
    (hidden
    0.11
    Activations Density 0.514%

    No Known Activations