© 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. 21661
    Prev
    Next
    INDEX
    Explanations

    presence of

    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
    å§Ķä¼ļ
    -0.10
    纵横
    -0.10
    éĵį
    -0.10
    沤
    -0.10
    swith
    -0.09
    ÑĢин
    -0.09
    æľ¬æĿ¥å°±
    -0.09
    çļĦè¿ĽæŃ¥
    -0.09
    èģĶåIJĪä¼ļ
    -0.09
    æľīæĿĥ
    -0.09
    POSITIVE LOGITS
    æĨ¾
    0.09
    nant
    0.09
    Ùģ
    0.09
    earch
    0.09
     налиÑĩии
    0.09
    à¸Ĺาà¸ĩ
    0.09
    ele
    0.08
     Ñģобой
    0.08
    çĻĸ
    0.08
    cript
    0.08
    Activations Density 0.039%

    No Known Activations