© 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. 52460
    Prev
    Next
    INDEX
    Explanations

    acts, act, or related words

    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
    ette
    -0.11
    hip
    -0.10
    æѤåĪ»
    -0.10
    ĭ
    -0.10
    ogle
    -0.10
    sic
    -0.10
    apolis
    -0.09
    ettes
    -0.09
    تÙĦ
    -0.09
    änd
    -0.09
    POSITIVE LOGITS
    uate
    0.18
    ually
    0.16
    ively
    0.16
    IVITY
    0.14
    iveness
    0.13
    uator
    0.13
    ives
    0.12
    rees
    0.12
    uates
    0.11
    UAL
    0.11
    Activations Density 0.066%

    No Known Activations