© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52182
    Prev
    Next
    INDEX
    Explanations

    fell short

    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
    ç¬ı
    -0.10
    æĬ¤èº«
    -0.10
    umerator
    -0.09
    ble
    -0.09
    åı°ä¸Ĭ
    -0.09
    .fd
    -0.08
    çĶŁäº§æĢ»åĢ¼
    -0.08
    樾
    -0.08
    èĩªæѤ
    -0.08
    aina
    -0.08
    POSITIVE LOGITS
     expectations
    0.10
    reach
    0.10
    åģļ个
    0.10
    é¢ĦæľŁ
    0.10
    åºĶæľīçļĦ
    0.10
    çIJĨæĥ³çļĦ
    0.09
     reality
    0.09
    ened
    0.09
    缮æłĩ
    0.08
    ioned
    0.08
    Activations Density 0.008%

    No Known Activations