© 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. 51516
    Prev
    Next
    INDEX
    Explanations

    names with 'il', 'ig', 'icit', 'ix', 'iz', 'is', 'ith' prefixes

    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
    еÑĢÑĮ
    -0.11
    åĴĮåľ°åĮº
    -0.11
    IJ
    -0.10
    çĴĥ
    -0.10
    Ä°
    -0.10
    åķ·
    -0.10
    ega
    -0.10
    elling
    -0.09
    .SYSTEM
    -0.09
    妣
    -0.09
    POSITIVE LOGITS
    zelf
    0.14
    rary
    0.14
    ê¹IJ
    0.12
    à¨Ĩ
    0.11
    ellaneous
    0.11
    keit
    0.11
    ãĥ§ãĥ³
    0.11
    çĽİ
    0.11
    ussion
    0.11
    olation
    0.11
    Activations Density 2.848%

    No Known Activations