© 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. 56032
    Prev
    Next
    INDEX
    Explanations

    initiating discussion or conversation

    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
     bis
    -0.09
    åĨįåĬłä¸Ĭ
    -0.09
    bis
    -0.09
    çĽ
    -0.09
     Bis
    -0.08
    赤
    -0.08
     Fro
    -0.08
    åĬ³
    -0.08
     Trot
    -0.08
    人工
    -0.08
    POSITIVE LOGITS
    çļĦè¯Ŀé¢ĺ
    0.23
    è¯Ŀé¢ĺ
    0.22
     conversation
    0.19
    è°Īè¯Ŀ
    0.16
     Conversation
    0.15
    conversation
    0.14
    è°Ī论
    0.14
    èģĬ
    0.14
     Gespräch
    0.14
     conversations
    0.13
    Activations Density 0.064%

    No Known Activations