© 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. 7679
    Prev
    Next
    INDEX
    Explanations

    retaliate, vindictive, persecuted

    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
    çĽ²çĽ®
    -0.10
    æĺ¼
    -0.09
    æĺİæľĹ
    -0.09
    ogy
    -0.09
    åıĮçľ¼
    -0.09
    äºĨåįĬ天
    -0.09
    èĪªæµ·
    -0.09
    æİıåĩº
    -0.08
    éŁ¿
    -0.08
    å¾ĴæŃ¥
    -0.08
    POSITIVE LOGITS
    æīĵåİĭ
    0.13
    诬
    0.12
    æĮĩæݧ
    0.11
    æĬ¥å¤į
    0.11
    åİĭåĪ¶
    0.10
     vind
    0.10
    æĮijåīĶ
    0.10
    vil
    0.10
     reflex
    0.09
    诽
    0.09
    Activations Density 0.106%

    No Known Activations