© 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. 8376
    Prev
    Next
    INDEX
    Explanations

    title separators

    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
    åĽ¢
    -0.10
    stå
    -0.09
    ä¸ĩéĩĮ
    -0.09
    åĽ°æī°
    -0.09
    æŦ
    -0.09
    äºī
    -0.09
    éĿŀ常çļĦ
    -0.08
    Ñģал
    -0.08
    åľĺ
    -0.08
     plat
    -0.08
    POSITIVE LOGITS
    asca
    0.10
    ãģ§ãģĤãĤįãģĨ
    0.08
    WSC
    0.08
    åĵªç§į
    0.08
    pai
    0.08
    QS
    0.08
     maxValue
    0.08
    eil
    0.08
    uhe
    0.08
    olla
    0.08
    Activations Density 0.101%

    No Known Activations