© 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. 10662
    Prev
    Next
    INDEX
    Explanations

    positive sentiment exclamations and endearments

    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
    æĤ¼
    -0.10
    èĢģçĪ·
    -0.10
    èĢģçĪ·åŃIJ
    -0.09
    çĦ¯
    -0.09
    _CRE
    -0.09
    é«Ń
    -0.09
    ä¸Ģçĵ¶
    -0.09
    roma
    -0.09
    dit
    -0.09
    odable
    -0.09
    POSITIVE LOGITS
    å®ķ
    0.11
    ä¸İåıijå±ķ
    0.09
     пен
    0.09
    æľĢçĪ±
    0.09
     Danger
    0.08
    Letter
    0.08
    lian
    0.08
     Letters
    0.08
    Danger
    0.08
    èį·
    0.08
    Activations Density 0.167%

    No Known Activations