© 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. 51569
    Prev
    Next
    INDEX
    Explanations

    brand new

    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
     Wonderland
    -0.09
     PROVIDED
    -0.09
    ç¼İ
    -0.08
    ird
    -0.08
    èijµ
    -0.08
    çĻ½éĽª
    -0.08
     Basel
    -0.08
    仪
    -0.08
    exion
    -0.08
     bamb
    -0.08
    POSITIVE LOGITS
    ishment
    0.10
    æĸ°ä¸ī
    0.09
    é¾
    0.09
    enburg
    0.09
    sar
    0.08
    igned
    0.08
    çĹĪ
    0.08
    æĪijèĥ½
    0.08
    anim
    0.08
    anc
    0.08
    Activations Density 0.020%

    No Known Activations