© 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. 21978
    Prev
    Next
    INDEX
    Explanations

    imperfect, rough, imp, hazards

    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
    æĪ¡
    -0.10
    好åIJĹ
    -0.10
    éĤ¸
    -0.10
    主ä¹īèĢħ
    -0.09
    èµĥ
    -0.09
    çļ¿
    -0.09
    ç½ħ
    -0.09
    åIJįçīĩ
    -0.09
    黥
    -0.09
    å¾Ĺ天
    -0.09
    POSITIVE LOGITS
    è·Ľ
    0.14
    -legged
    0.11
    çĺ¸
    0.10
    æ¯ĽçĹħ
    0.10
    éļIJ
    0.09
    è¡Įèµ°
    0.09
    è¸ī
    0.09
    èµ°è·¯
    0.08
    à¹Ħหล
    0.08
    æĬĸ
    0.08
    Activations Density 0.053%

    No Known Activations