© 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. 8175
    Prev
    Next
    INDEX
    Explanations

    Square names

    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
     survivor
    -0.09
    åĮħ袱
    -0.09
    大æĪ·
    -0.08
    åĪ¹
    -0.08
    æ±¾
    -0.08
    ĥģ
    -0.08
    æĨij
    -0.08
    èĢĮä¸Ķè¿ĺ
    -0.08
    som
    -0.08
    åĩŃ
    -0.08
    POSITIVE LOGITS
    ä¸ĭåij¨
    0.10
    ilet
    0.10
    æľ¬åij¨
    0.09
     ÑģÑĤоиÑĤ
    0.09
    upp
    0.09
    enta
    0.09
    èĽĭç³ķ
    0.09
    èĪŀ
    0.09
    /Area
    0.09
    pets
    0.09
    Activations Density 0.010%

    No Known Activations