© 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. 10427
    Prev
    Next
    INDEX
    Explanations

    excuse to justify attack

    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
    ãģĶãģĸ
    -0.10
    æķĻæĿIJ
    -0.09
    åįķåįķ
    -0.09
    6
    -0.08
    çķľ
    -0.08
    DataService
    -0.08
     ISO
    -0.08
    çIJĩ
    -0.08
    éĻ¤éĿŀ
    -0.08
    è½®æµģ
    -0.08
    POSITIVE LOGITS
    è·ĭ
    0.11
    å±ģ
    0.10
    é£İ鼨
    0.10
    æīĢéľĢè¦ģ
    0.10
    yles
    0.10
     rhyme
    0.09
    .ua
    0.09
    å¥ĭ
    0.09
    æīĢéľĢè¦ģçļĦ
    0.09
    etto
    0.09
    Activations Density 0.064%

    No Known Activations