© 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. 22649
    Prev
    Next
    INDEX
    Explanations

    past, before, previously

    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
    æ¯ı天éĥ½
    -0.09
    ä¾Ŀæŧæĺ¯
    -0.09
    åłĨ积
    -0.09
    媳
    -0.09
    èµ·åĪĿ
    -0.09
     somehow
    -0.08
     initially
    -0.08
    è¯ļå®ŀ
    -0.08
     latest
    -0.08
    èĪ¶
    -0.08
    POSITIVE LOGITS
    å¤ļ次
    0.13
    说è¿ĩ
    0.12
    æľīè¿ĩ
    0.11
    ä¸įæŃ¢
    0.11
    ä»ĸæĽ¾
    0.11
    око
    0.10
    åħ¶ä»ĸ
    0.10
     elsewhere
    0.10
    oca
    0.09
    stroy
    0.09
    Activations Density 0.170%

    No Known Activations