© 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. 63062
    Prev
    Next
    INDEX
    Explanations

    ter 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
    éri
    -0.10
    urations
    -0.10
    ellers
    -0.09
    eries
    -0.09
    ÑĢай
    -0.09
    astic
    -0.09
    åģıåģı
    -0.09
    æµ®çİ°
    -0.09
    tas
    -0.08
    åĥ®
    -0.08
    POSITIVE LOGITS
    prising
    0.11
    TING
    0.11
    riers
    0.10
    ủng
    0.10
    rible
    0.10
    usan
    0.10
    woord
    0.10
    ÙĬض
    0.10
    riere
    0.09
    iously
    0.09
    Activations Density 0.036%

    No Known Activations