© 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. 26915
    Prev
    Next
    INDEX
    Explanations

    most + follow-up adjective

    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
     sar
    -0.08
    URI
    -0.08
    éĻŁ
    -0.08
    éģĹçķĻ
    -0.08
    éĿ¢åīį
    -0.08
    ahl
    -0.08
    dit
    -0.08
     SAR
    -0.07
    anel
    -0.07
    æŃ£å¸¸çļĦ
    -0.07
    POSITIVE LOGITS
    æľĢå¤ļçļĦ
    0.17
    æľĢå¤ļ
    0.17
    æ¯Ķè¾ĥå¤ļ
    0.15
    è¾ĥå¤ļ
    0.15
    è¶Ĭå¤ļ
    0.15
    æľĢ为
    0.15
    å¾Ĺå¤ļ
    0.12
    æľĢåıĹ
    0.12
    è¶Ĭ大
    0.11
    尤为
    0.11
    Activations Density 0.066%

    No Known Activations