© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 37679
    Prev
    Next
    INDEX
    Explanations

    is the

    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
    éĵ¨
    -0.09
    å°Ħ
    -0.09
    itter
    -0.09
    yo
    -0.09
    roup
    -0.09
    íĨµ
    -0.09
    ä¸ī代
    -0.08
     rej
    -0.08
    ól
    -0.08
     shoot
    -0.08
    POSITIVE LOGITS
     Affero
    0.10
    æ²ī
    0.09
    quals
    0.09
    èŀĪ
    0.09
    æīĨ
    0.09
    ä¸ĢåĪ»
    0.09
    ç¢İ
    0.09
    ÑĨеп
    0.09
    æ¯ķä¸ļåIJİ
    0.08
     Nb
    0.08
    Activations Density 0.093%

    No Known Activations