© 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. 37238
    Prev
    Next
    INDEX
    Explanations

    chair and its context

    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
    ted
    -0.12
    imonial
    -0.11
    rog
    -0.10
    ÑģÑı
    -0.09
    oses
    -0.09
    è¿Ļ家
    -0.09
    æĸ°ä¸Ģ代
    -0.09
    çıŃ
    -0.09
    sect
    -0.09
    çĵ¶
    -0.09
    POSITIVE LOGITS
    ulin
    0.10
    apult
    0.09
    åĽ°å¢ĥ
    0.08
    å²Ń
    0.08
    unas
    0.08
    anka
    0.08
    acters
    0.08
    MetaData
    0.08
     comforts
    0.08
    èĤ¯å®ļæĺ¯
    0.08
    Activations Density 0.010%

    No Known Activations