© 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. 58294
    Prev
    Next
    INDEX
    Explanations

    but / however conjunctions

    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
    æĪij没
    -0.12
    æĪij没æľī
    -0.11
    ç»Ļèĩªå·±
    -0.10
    edin
    -0.10
    æ¯ĶæĪij
    -0.10
    æŃ£å¸¸ä½¿çĶ¨
    -0.09
    èĻ
    -0.09
    heimer
    -0.09
    å¾ĪæŃ£å¸¸
    -0.09
    ä¸įç®Ĺ
    -0.09
    POSITIVE LOGITS
    æŀ¶
    0.09
    åį´
    0.09
    wel
    0.09
    åį´åıijçİ°
    0.09
    irts
    0.08
    ZE
    0.08
    HING
    0.08
    extr
    0.08
    eos
    0.08
    ween
    0.08
    Activations Density 0.225%

    No Known Activations