© 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. 56978
    Prev
    Next
    INDEX
    Explanations

    common conjunctions followed by verbs

    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
    æĥ°
    -0.09
    çĸ¾
    -0.09
    æĢİä¹Īèĥ½
    -0.08
    accessible
    -0.08
    ires
    -0.08
    onor
    -0.08
    ä¸ĢèĪ¬çļĦ
    -0.08
    åħħ足çļĦ
    -0.08
    yle
    -0.08
    çļĦå½¢å¼ı
    -0.08
    POSITIVE LOGITS
    æĺ¯åĽłä¸º
    0.17
    è¿Ļå°±æĺ¯
    0.17
    éĤ£æĺ¯
    0.17
    éĤ£å°±æĺ¯
    0.17
    è¿Ļæĺ¯åĽłä¸º
    0.16
    å°±æĺ¯åĽłä¸º
    0.15
    è¿ĻäºĽéĥ½æĺ¯
    0.15
    å®Łãģ¯
    0.15
    åħ¶å®ŀæĺ¯
    0.15
    åħ¶å®ŀå°±æĺ¯
    0.14
    Activations Density 0.111%

    No Known Activations