© 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. 22517
    Prev
    Next
    INDEX
    Explanations

    emergent levels and transitions

    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
     Jet
    -0.11
    éŸ
    -0.09
    å°±è·Ł
    -0.09
    åĵ²
    -0.09
    åĽ¾çīĩæĿ¥æºIJ
    -0.09
    å¤©åĽ½
    -0.09
    oga
    -0.09
    ç»ŀ
    -0.09
    å®ŀè´¨
    -0.08
     jet
    -0.08
    POSITIVE LOGITS
     emerg
    0.10
     divorced
    0.09
    å®ıè§Ĥ
    0.09
    稳å®ļçļĦ
    0.09
     Weiner
    0.09
    adle
    0.09
     tä
    0.08
    å®ıè§Ĥç»ıæµİ
    0.08
    è§Ĥ
    0.08
     emergence
    0.08
    Activations Density 0.085%

    No Known Activations