© 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. 36826
    Prev
    Next
    INDEX
    Explanations

    violet, violator, violence

    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
    éģĵè·¯ä¸Ĭ
    -0.09
    epy
    -0.09
     quo
    -0.09
    wick
    -0.09
    د
    -0.09
    orum
    -0.09
    reen
    -0.09
    æľĢåĸľæ¬¢çļĦ
    -0.09
    chaft
    -0.08
    æĬµåĪ¶
    -0.08
    POSITIVE LOGITS
    olle
    0.09
    ncia
    0.08
    ience
    0.08
    enze
    0.08
    ance
    0.08
    assin
    0.08
    marsh
    0.08
    alty
    0.08
    omid
    0.08
    ange
    0.08
    Activations Density 0.007%

    No Known Activations