© 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. 36671
    Prev
    Next
    INDEX
    Explanations

    notably following word

    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
    æĢ§æĦŁ
    -0.09
    大èĩ´
    -0.09
    eners
    -0.09
    çĤ¹æ»´
    -0.08
    wares
    -0.08
    uku
    -0.08
    æľĶ
    -0.08
    agos
    -0.08
     доÑģÑĤÑĥп
    -0.08
    å¿»
    -0.08
    POSITIVE LOGITS
    åŀĤ
    0.10
    ably
    0.10
    ä¹ĭå¤Ħ
    0.10
    ville
    0.09
    ingly
    0.09
    ç¼
    0.09
     Dunn
    0.09
    ãĥ¼ãĥ³
    0.09
    ä¾ĭå¤ĸ
    0.09
    prs
    0.08
    Activations Density 0.024%

    No Known Activations