© 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. 21059
    Prev
    Next
    INDEX
    Explanations

    instead of / rather than

    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
    avit
    -0.09
     aj
    -0.09
    äºĨèµ·æĿ¥
    -0.08
    éģĩåĪ°äºĨ
    -0.08
    èİŀ
    -0.08
    æĪĸå¤ļæĪĸå°ij
    -0.08
     liá»ĩu
    -0.08
    åı¦æľī
    -0.08
    èµ·æĿ¥
    -0.08
    åĨı
    -0.08
    POSITIVE LOGITS
    象
    0.11
    殿åłĤ
    0.10
     instead
    0.10
    笼
    0.09
     Instead
    0.09
    обоÑĢоÑĤ
    0.09
    instead
    0.09
    statt
    0.09
    arus
    0.09
    Instead
    0.08
    Activations Density 0.028%

    No Known Activations