© 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. 10987
    Prev
    Next
    INDEX
    Explanations

    possessive words and contractions

    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
    个人è§ĤçĤ¹
    -0.11
    æīĢåľ¨
    -0.10
    å«Ĥ
    -0.09
    æĹ¶ä»£ä¸ŃåĽ½
    -0.09
    _GP
    -0.08
    åīįæĸ¹
    -0.08
    æĺ¯ä¸Ģä½į
    -0.08
    å·¥ä½ľç»ıéªĮ
    -0.08
    å¿ĥæĦı
    -0.08
    aped
    -0.08
    POSITIVE LOGITS
    微信åħ¬ä¼Ĺ
    0.12
    èī¯å¥½
    0.10
    Öģ
    0.09
    äºī
    0.09
    çķ¸
    0.09
    éĤ£æ¬¡
    0.08
     <->
    0.08
    _CTL
    0.08
    inho
    0.08
    èµľ
    0.08
    Activations Density 0.165%

    No Known Activations