© 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. 2371
    Prev
    Next
    INDEX
    Explanations

    coral reefs

    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
    é¼¹
    -0.10
    åIJķ
    -0.09
    åĩĽ
    -0.09
    ris
    -0.09
     MES
    -0.09
    æĶ¥
    -0.08
    RID
    -0.08
     LSB
    -0.08
    缪
    -0.08
    æŃ¤äºº
    -0.08
    POSITIVE LOGITS
     Reef
    0.10
     reefs
    0.10
     reef
    0.10
     Resort
    0.09
    èݲèĬ±
    0.09
    çıĬçijļ
    0.09
     coral
    0.09
    hone
    0.09
    built
    0.08
    åįİ侨
    0.08
    Activations Density 0.036%

    No Known Activations