© 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. 19437
    Prev
    Next
    INDEX
    Explanations

    will/would followed by a verb

    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
    çĽ²
    -0.10
    åĪĹåħ¥
    -0.10
    oned
    -0.09
    æĸ¹æīį
    -0.09
     upp
    -0.09
     ÑģейÑĩаÑģ
    -0.09
    æĪIJéķ¿ä¸º
    -0.09
    æĺ§
    -0.09
     поÑģледние
    -0.09
    plusplus
    -0.08
    POSITIVE LOGITS
    å±ĬæŶ
    0.09
    çļĦåħ³éĶ®
    0.09
    åįıä¼ļä¼ļåijĺ
    0.09
    á»±
    0.09
    å°Ĩæĺ¯
    0.09
    æŃĮæīĭ
    0.08
    å·¥ç¨ĭæĬĢæľ¯
    0.08
    -available
    0.08
    competition
    0.08
     bÄĻdzie
    0.08
    Activations Density 0.090%

    No Known Activations