© 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. 20968
    Prev
    Next
    INDEX
    Explanations

    numbers like 6, 360, 65

    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
     ÑģобÑģÑĤвенно
    -0.09
    ugs
    -0.08
    ãģ©
    -0.08
    饶
    -0.08
    Ñıд
    -0.08
    åĪij
    -0.08
     compromise
    -0.08
    éĹŃ
    -0.08
     Schwe
    -0.07
    hid
    -0.07
    POSITIVE LOGITS
    èIJ¥ä¸ļ
    0.10
    ç´§æī£
    0.10
    羸
    0.10
    alls
    0.10
    è§Ĥæµĭ
    0.09
    åıĹå½±åĵį
    0.09
    é¹¢
    0.09
     phú
    0.09
    åĿĬ
    0.09
     Daly
    0.09
    Activations Density 0.025%

    No Known Activations