© 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. 8297
    Prev
    Next
    INDEX
    Explanations

    following conditions

    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
    æĺ¯ä¸Ģ件
    -0.09
    æĹĥ
    -0.09
    è¿Ļ个éĹ®é¢ĺ
    -0.09
    ients
    -0.09
    麦åħĭ
    -0.08
    鼷æĸ¯
    -0.08
    tridges
    -0.08
    .hw
    -0.08
    åĹĵ
    -0.08
    akh
    -0.08
    POSITIVE LOGITS
    ä¸Ģ个个
    0.15
    åIJĦåľ°
    0.12
     коÑĤоÑĢÑĭе
    0.12
    оÑĩнÑĭе
    0.11
     ÑĤакие
    0.10
     Twin
    0.10
     коÑĤоÑĢÑĭÑħ
    0.10
    åīįåIJİ
    0.10
    诸å¦Ĥ
    0.10
    åIJĦ个
    0.09
    Activations Density 0.820%

    No Known Activations