© 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. 11464
    Prev
    Next
    INDEX
    Explanations

    had/is/was/are + past participle or 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
    å¯Ħ
    -0.11
    åºĦ严
    -0.10
    ç¿»
    -0.10
     ведÑĮ
    -0.09
    adesh
    -0.09
    称ä¹ĭ为
    -0.09
    _INFINITY
    -0.09
    ismet
    -0.09
    被称为
    -0.09
    .Navigator
    -0.09
    POSITIVE LOGITS
    æľīæľĽ
    0.09
    ãĤ£
    0.08
    çī²
    0.08
     difference
    0.08
    ä¸įåıªæĺ¯
    0.08
    åĪĿæŃ¥
    0.08
    ãģłãģĮ
    0.08
    ê¹IJ
    0.07
    åīįæĻ¯
    0.07
    è¦ģä¹Ī
    0.07
    Activations Density 0.088%

    No Known Activations