© 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. 8026
    Prev
    Next
    INDEX
    Explanations

    statements about assumptions

    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
    æĽ²æĬĺ
    -0.12
    æıŃ
    -0.08
    æ·¦
    -0.08
    ays
    -0.08
    ä¸Ģéĥ¨åĪĨ
    -0.08
    éĺĥ
    -0.08
    çİ°æľī
    -0.08
    marvin
    -0.08
     ske
    -0.08
    ãģĦãģıãĤī
    -0.08
    POSITIVE LOGITS
     assumptions
    0.25
    åģĩ设
    0.21
     assumption
    0.20
    åīįæıIJ
    0.19
     assum
    0.18
    çļĦåīįæıIJ
    0.15
    åīįæıIJæĺ¯
    0.15
     assumes
    0.15
     assumed
    0.14
     assume
    0.14
    Activations Density 0.083%

    No Known Activations