© 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. 53035
    Prev
    Next
    INDEX
    Explanations

    numbers in the millions

    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
     Kirk
    -0.09
    åIJķ
    -0.09
     Vig
    -0.09
     Flames
    -0.08
     Bulls
    -0.08
    ãģĵãģĵãģ¾ãģ§
    -0.08
    Cors
    -0.08
     Ultimate
    -0.08
     Prest
    -0.08
     Pan
    -0.07
    POSITIVE LOGITS
    åŃĻåŃIJ
    0.11
     Wallace
    0.10
    æį¶
    0.10
    ç©°
    0.09
    è¾ľè´Ł
    0.09
    å°ıåŃIJ
    0.09
    éľ²å¤©
    0.09
     baj
    0.09
    æµģä¼ł
    0.09
    就好
    0.09
    Activations Density 0.066%

    No Known Activations