© 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. 19881
    Prev
    Next
    INDEX
    Explanations

    agile, dexterity

    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
    滥
    -0.09
    æ½®
    -0.09
    Hack
    -0.09
     Canyon
    -0.09
    idal
    -0.09
    ollider
    -0.08
     vigil
    -0.08
    uro
    -0.08
    åĬĽæ±Ĥ
    -0.08
    åĿļå®ļä¸į移
    -0.08
    POSITIVE LOGITS
    æĹłæ¯Ķ
    0.12
    NESS
    0.11
     enough
    0.10
     Minds
    0.10
    æįķæįī
    0.10
    æĿ¼
    0.09
    èĩªå¦Ĥ
    0.09
    shal
    0.09
     ens
    0.08
    ä½İä¸ĭ
    0.08
    Activations Density 0.007%

    No Known Activations