© 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. 3806
    Prev
    Next
    INDEX
    Explanations

    planning, proposals, intentions

    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
    è´Ŀ
    -0.10
    äºĨ解åĪ°
    -0.09
    åīįåĪĹ
    -0.08
    ä¸ĭåįĪ
    -0.08
    没èĥ½
    -0.08
    ç«Ń
    -0.08
    PD
    -0.08
    ç¯ĩ
    -0.08
    误解
    -0.08
    leep
    -0.08
    POSITIVE LOGITS
    cht
    0.10
    æīĭæĮģ
    0.10
    ä¸Ģæīĭ
    0.09
    åıĹ访
    0.09
     Sanford
    0.09
    åīįæıIJæĺ¯
    0.09
    çĭ¬ç«ĭèij£äºĭ
    0.09
    roscope
    0.09
    缸符
    0.09
    oland
    0.09
    Activations Density 0.113%

    No Known Activations