© 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. 36019
    Prev
    Next
    INDEX
    Explanations

    sy words and suffixes

    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
    å§Ŀ
    -0.10
    åħħå®ŀ
    -0.10
    azy
    -0.10
    ï¸ı
    -0.09
    ci
    -0.09
    _COMPLEX
    -0.09
    èĪģ
    -0.09
    eur
    -0.09
    ç¬Ļ
    -0.09
    suppress
    -0.08
    POSITIVE LOGITS
    è¨Ĺ
    0.11
    onym
    0.10
    NTAX
    0.10
    æĢ§å¼º
    0.10
    onymous
    0.10
    MBOL
    0.08
    onyms
    0.08
    onomy
    0.08
    ached
    0.08
    dney
    0.08
    Activations Density 0.035%

    No Known Activations