© 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. 61932
    Prev
    Next
    INDEX
    Explanations

    how long, how many, what, which

    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¢ŀéĩı
    -0.10
     sounding
    -0.10
    æĪijåıijçİ°
    -0.09
    ä¸įçķĻ
    -0.09
    管çIJĨå·¥ä½ľ
    -0.09
     Juice
    -0.09
    æĪijä»¬ä¹Ł
    -0.09
    主ä¸ļ
    -0.09
    å·§
    -0.09
    çĽĸ
    -0.08
    POSITIVE LOGITS
    ä½ķæŶ
    0.12
    éĢłèĪ¹
    0.10
    è¿Ľå±ķæĥħåĨµ
    0.10
    راء
    0.10
    æĺ¯åIJ¦
    0.09
    è¿Ľåº¦
    0.09
     progress
    0.09
    åıijå±ķè¶ĭåĬ¿
    0.09
    èĩªæĿ¥
    0.09
    agan
    0.09
    Activations Density 0.185%

    No Known Activations