© 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. 48152
    Prev
    Next
    INDEX
    Explanations

    any damages or purpose

    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
    纤
    -0.09
    åİĭ
    -0.08
    ãģĻãģ¹
    -0.08
    ãģŃãģĩ
    -0.08
    она
    -0.08
    iron
    -0.08
    ëıĪ
    -0.08
     salvage
    -0.08
    redd
    -0.08
    3
    -0.07
    POSITIVE LOGITS
    æ±Ĺ
    0.10
    泺
    0.09
    mı
    0.09
    èľ¥
    0.09
    社ä¼ļ稳å®ļ
    0.09
    .esp
    0.09
    IZES
    0.09
    -gap
    0.09
    æĢĿç»´æĸ¹å¼ı
    0.09
     çıkan
    0.09
    Activations Density 0.012%

    No Known Activations