© 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. 56732
    Prev
    Next
    INDEX
    Explanations

    merits, when, regardless, identification, gap

    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
    å´Ĩ
    -0.10
     somew
    -0.10
    æłĩåĩĨåĮĸ
    -0.08
    å¯ĵ
    -0.08
    ynom
    -0.08
    èģĶ绾
    -0.08
    鲤
    -0.08
    -marker
    -0.08
    å¨Ľæ¨Ĥ
    -0.08
    reon
    -0.08
    POSITIVE LOGITS
    èĢĮä¸įæĺ¯
    0.11
    éģ´
    0.10
     merit
    0.10
    èĢĮéĿŀ
    0.09
    resse
    0.09
     criteria
    0.09
     Neal
    0.09
    ä»»
    0.09
     lá»Ļ
    0.09
    criteria
    0.09
    Activations Density 0.061%

    No Known Activations