© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37123
    Prev
    Next
    INDEX
    Explanations

    all followed by another word

    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
    ä»İ严
    -0.09
     Memo
    -0.08
    ppers
    -0.08
    åIJģ
    -0.08
     fed
    -0.08
    ç©´
    -0.08
     обÑıзаÑĤелÑĮно
    -0.08
    æĿ¯
    -0.08
    åħ®
    -0.08
    é«ĺä»·
    -0.08
    POSITIVE LOGITS
    az
    0.09
    iat
    0.09
    认è¯ģ
    0.09
     Enumerator
    0.09
    æĹºçĽĽ
    0.09
    INGER
    0.09
    ajes
    0.08
    ém
    0.08
    è¶ĭåĬ¿
    0.08
    anted
    0.08
    Activations Density 0.035%

    No Known Activations