© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 17033
    Prev
    Next
    INDEX
    Explanations

    courtyard and yard descriptions

    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
    溺
    -0.09
    绸
    -0.09
    çĥ¦æģ¼
    -0.09
    èį§
    -0.09
    åĽ½å®¶æĪĺçķ¥
    -0.09
    (&(
    -0.09
     Cave
    -0.09
     cave
    -0.09
    åĩŃ
    -0.09
    auge
    -0.09
    POSITIVE LOGITS
    éĻ¢åŃIJéĩĮ
    0.14
    éĻ¢åŃIJ
    0.11
    该éĻ¢
    0.11
    éĻ¢
    0.11
     courtyard
    0.10
    å¹¿åľº
    0.10
    è¯¥æł¡
    0.10
    æł¡åĽŃ
    0.09
    åºŃ
    0.09
    รม
    0.09
    Activations Density 0.101%

    No Known Activations