© 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. 31822
    Prev
    Next
    INDEX
    Explanations

    past tense verb + describing outcome

    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
    addtogroup
    -0.09
    azzi
    -0.09
    MEA
    -0.09
    èĤĦ
    -0.09
     arab
    -0.09
    usan
    -0.09
     Carlton
    -0.09
    ä¸Ĭ课
    -0.09
    大ä¼ļä¸Ĭ
    -0.09
    æ¯ģçģŃ
    -0.09
    POSITIVE LOGITS
    ç»ĵæĿŁåIJİ
    0.14
    åıijçĶŁåľ¨
    0.13
    ç»ĵæĿŁ
    0.13
    ç»ĵæĿŁäºĨ
    0.11
    çµIJæĿŁ
    0.10
    å®īæİĴ
    0.10
     conducted
    0.10
     diá»ħn
    0.10
    èIJ½å¹ķ
    0.10
    åħ¨ç¨ĭ
    0.10
    Activations Density 0.233%

    No Known Activations