© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8218
    Prev
    Next
    INDEX
    Explanations

    play followed by gaming or media terms

    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
    nement
    -0.11
    ized
    -0.11
    bons
    -0.10
    ised
    -0.10
    /from
    -0.10
    ials
    -0.10
    ities
    -0.09
    chin
    -0.09
    ëĿ¼ëıĦ
    -0.09
    inals
    -0.09
    POSITIVE LOGITS
    ings
    0.13
    wright
    0.12
    fully
    0.12
     havoc
    0.11
    manship
    0.10
    ERS
    0.10
    backs
    0.09
    çħ
    0.09
    ables
    0.09
    offs
    0.09
    Activations Density 0.065%

    No Known Activations