© 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. 51757
    Prev
    Next
    INDEX
    Explanations

    upholstery, furniture, bind, wall, holstered

    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¤ĩæ¡Ī
    -0.09
    isco
    -0.08
    cale
    -0.08
    igo
    -0.08
    olley
    -0.08
    项éĵ¾
    -0.08
    èĪ¶
    -0.08
    -eyed
    -0.08
    ä¸Ĭåįĥ
    -0.08
     обÑĢазом
    -0.08
    POSITIVE LOGITS
    ment
    0.09
    æ¾ľ
    0.09
    hung
    0.09
    人åijĺ
    0.09
    ieri
    0.09
    çĤķ
    0.09
    ier
    0.09
    unker
    0.08
    å¼Ĭ
    0.08
    enting
    0.08
    Activations Density 0.007%

    No Known Activations