© 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. 39043
    Prev
    Next
    INDEX
    Explanations

    halogens, pseudohalogens

    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
    ÙĨج
    -0.10
    å¼ĢèįĴ
    -0.09
    夫人
    -0.08
    éķ¿æ±Ł
    -0.08
     Singh
    -0.08
    æľīæĦı
    -0.08
     Wheeler
    -0.08
     Scholars
    -0.08
    ä»İ严
    -0.07
     Clement
    -0.07
    POSITIVE LOGITS
    åĢŃ
    0.10
    æĭĴ
    0.10
    zon
    0.09
    訾
    0.09
    ries
    0.09
    ÑĽ
    0.09
    idth
    0.09
    è·Łéļı
    0.09
    å®īåħ¨ä¿Ŀéļľ
    0.09
    ä¸įæ¸ħæ¥ļ
    0.09
    Activations Density 0.003%

    No Known Activations