© 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. 38300
    Prev
    Next
    INDEX
    Explanations

    across regions 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
    å±¥
    -0.11
    åĤĢ
    -0.10
     Accessed
    -0.09
    ä»Ģä¹Īæł·çļĦ
    -0.09
    ť
    -0.09
    бо
    -0.09
    éĵº
    -0.09
    åħ¬ç¤º
    -0.09
     Fitz
    -0.08
    åĮ¿
    -0.08
    POSITIVE LOGITS
    è·¨
    0.26
    è·¨è¶Ĭ
    0.20
    è·¨çķĮ
    0.15
     across
    0.14
     boundaries
    0.13
    Across
    0.13
     cross
    0.12
     Across
    0.11
    ç©¿è¶Ĭ
    0.11
    cross
    0.11
    Activations Density 0.155%

    No Known Activations