© 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. 61126
    Prev
    Next
    INDEX
    Explanations

    regular or regularly followed by noun

    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
    -0.11
    /her
    -0.10
    лем
    -0.10
    arel
    -0.10
    rial
    -0.10
    raid
    -0.09
    icial
    -0.09
    رÙĪب
    -0.09
    ละ
    -0.09
    çĦ¶æĺ¯
    -0.09
    POSITIVE LOGITS
    ities
    0.18
    ity
    0.17
    ITY
    0.13
    ised
    0.12
    iser
    0.12
    mente
    0.12
     xuyên
    0.12
    ONTAL
    0.11
    dehyde
    0.11
    izedName
    0.11
    Activations Density 0.027%

    No Known Activations