© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 10048
    Prev
    Next
    INDEX
    Explanations

    despite..." contrast

    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
    å²Ĥ
    -0.10
     знаÑĩиÑĤ
    -0.10
     Saud
    -0.10
    çªģçł´åı£
    -0.10
    ç§ķ
    -0.09
    太å¤ļäºĨ
    -0.09
    ä¸į该
    -0.09
    å¹²èĦĨ
    -0.09
    one
    -0.09
     tháºŃm
    -0.08
    POSITIVE LOGITS
    è¿Ļä¸ĢçĤ¹
    0.13
    è¿ĻçĤ¹
    0.12
    ively
    0.11
     ë¶Ī구íķĺê³ł
    0.11
     having
    0.11
    emente
    0.11
     being
    0.11
     its
    0.09
     ÑĤого
    0.09
    ç§įç§į
    0.09
    Activations Density 0.018%

    No Known Activations