© 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. 51734
    Prev
    Next
    INDEX
    Explanations

    equally followed by adjective/adverb

    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
    icol
    -0.10
    agation
    -0.09
    umb
    -0.08
    ught
    -0.08
     same
    -0.08
    ussy
    -0.08
    gether
    -0.08
    aha
    -0.08
    irectory
    -0.08
    ç¬ijäºĨ
    -0.07
    POSITIVE LOGITS
    mente
    0.11
    -sized
    0.11
    档次
    0.11
    emente
    0.10
     ë¶Ī구íķĺê³ł
    0.10
    åı¤èĢģ
    0.10
    aneous
    0.10
    iy
    0.10
    ately
    0.10
     sized
    0.10
    Activations Density 0.007%

    No Known Activations