© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 33121
    Prev
    Next
    INDEX
    Explanations

    groups of two, three, or four

    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
    rots
    -0.09
    ÑĨов
    -0.09
    æijħ
    -0.09
    éĨ®
    -0.09
    åĬłåĪĨ
    -0.09
    æī«ä¸Ģæī«
    -0.09
    oras
    -0.09
    ObjectOfType
    -0.09
    .ColumnHeader
    -0.09
    éĤ¨
    -0.09
    POSITIVE LOGITS
    ä¸ī人
    0.16
    å¤ļ人
    0.12
    äºĮ人
    0.11
    两人
    0.11
    人æķ°
    0.10
    è¿ĻäºĽäºº
    0.10
    åĩłä¸ªäºº
    0.10
    两个人
    0.10
    仨
    0.09
    ä¿©
    0.09
    Activations Density 0.146%

    No Known Activations