© 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. 50962
    Prev
    Next
    INDEX
    Explanations

    punctuation followed by capitalized words

    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
    çīł
    -0.09
    æĢ§åij½
    -0.09
    å·¾
    -0.09
     ped
    -0.08
    NES
    -0.08
     [âĢ¦]
    -0.08
    âĢ¤
    -0.08
    Ìģ
    -0.08
    .TryParse
    -0.07
    ç¥Ń
    -0.07
    POSITIVE LOGITS
    è¦ģçľĭ
    0.11
    第ä¸ĢåIJį
    0.10
    人群
    0.09
    åIJĮæł·æĺ¯
    0.09
    å¾Ħ
    0.09
    åĨįéĢł
    0.09
    shaw
    0.09
    åĨį度
    0.09
    insic
    0.09
    好åIJĹ
    0.09
    Activations Density 0.103%

    No Known Activations