© 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
    2. Qwen3-32B
    3. 32-RESID-BATCHTOPK-65K
    4. 20869
    Prev
    Next
    INDEX
    Explanations

    old-school or old-fashioned

    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
    atto
    -0.09
    little
    -0.09
    åīįä¸ĸ
    -0.08
     Little
    -0.08
    nst
    -0.08
    éĩijåĪļ
    -0.08
    åı»
    -0.08
    åīįæīĢ
    -0.07
    -maker
    -0.07
    nce
    -0.07
    POSITIVE LOGITS
    ì§Ī
    0.11
    åħ¶ä½Ļ
    0.10
    è¾Ĺ
    0.10
    äºĤ
    0.09
    ä¹ĭä½ľ
    0.09
    ä¹±
    0.09
    ä»Ĺ
    0.09
    ording
    0.09
    éĸ£
    0.09
    峡
    0.09
    Activations Density 0.003%

    No Known Activations