© 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. 12008
    Prev
    Next
    INDEX
    Explanations

    still loves you

    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
    åĭĭç«ł
    -0.10
    åĶ¤éĨĴ
    -0.09
    æįķ
    -0.09
    å®īæĬļ
    -0.08
    æįķæįī
    -0.08
    人éĢī
    -0.08
    ULO
    -0.08
    æĴ·
    -0.08
    ration
    -0.08
    为äºĨ让
    -0.08
    POSITIVE LOGITS
    继ç»Ń
    0.16
    ä»į
    0.14
    ç¹¼çºĮ
    0.13
    éĩįè¿Ķ
    0.12
    ä»įæľī
    0.12
    ä¿ĿæĮģçĿĢ
    0.12
    继ç»Ńä¿ĿæĮģ
    0.12
     continued
    0.11
    ä»įçĦ¶
    0.11
    ä¿ĿçķĻ
    0.11
    Activations Density 0.187%

    No Known Activations