© 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. 21937
    Prev
    Next
    INDEX
    Explanations

    universities and research centers

    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
    ellig
    -0.09
    åŁĭ
    -0.09
    éĢģä¸Ĭ
    -0.09
    лов
    -0.09
    æ®Ħ
    -0.09
    ellers
    -0.09
    _ctxt
    -0.09
    inda
    -0.09
    rious
    -0.08
    ä¸Ģåı£
    -0.08
    POSITIVE LOGITS
    çĸ£
    0.10
     Cr
    0.09
     Cata
    0.08
    ower
    0.08
    壤
    0.08
    ono
    0.08
     اÙĦتاب
    0.07
    éªij士
    0.07
    announce
    0.07
    笤
    0.07
    Activations Density 0.043%

    No Known Activations