© 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. 20124
    Prev
    Next
    INDEX
    Explanations

    wooded areas or populated places

    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
    itt
    -0.08
    iest
    -0.08
    Ä«
    -0.08
    æŃ£çīĪ
    -0.08
    ster
    -0.08
    maid
    -0.08
    sic
    -0.08
    fold
    -0.08
    ist
    -0.08
    uliar
    -0.07
    POSITIVE LOGITS
    è¿ĻäºĽéĥ½æĺ¯
    0.11
    æ±ł
    0.11
     areas
    0.10
    大åİħ
    0.10
    åĮº
    0.10
    usto
    0.09
    ouser
    0.09
    ç»ıå¼ĢåĮº
    0.09
    åľ°åĮº
    0.09
    åĮºéĹ´
    0.09
    Activations Density 0.089%

    No Known Activations