© 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. 20391
    Prev
    Next
    INDEX
    Explanations

    stationary head, base, type, phase

    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
    odia
    -0.10
    igator
    -0.10
    人éĢł
    -0.09
    ãģŀ
    -0.09
    otics
    -0.09
    allen
    -0.09
    usi
    -0.08
     Fargo
    -0.08
    icket
    -0.08
    沧å·ŀ
    -0.08
    POSITIVE LOGITS
    ä¹ĭ乡
    0.10
    ç½Ĺæĸ¯
    0.09
    æģ£
    0.09
    ForResult
    0.09
    ç±į
    0.09
    帷
    0.09
    订è´Ń
    0.09
    å»ºåĽ½
    0.09
    åħħ足
    0.09
    çĬĦ
    0.09
    Activations Density 0.008%

    No Known Activations