© 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. 51809
    Prev
    Next
    INDEX
    Explanations

    started an action or activity

    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
    å·§åIJĪ
    -0.11
    èı½
    -0.09
    _dtype
    -0.08
    æ¶Ī
    -0.08
    iesen
    -0.08
    天çĦ¶
    -0.08
     Ñģебе
    -0.08
    woord
    -0.08
    溥
    -0.08
     Joan
    -0.08
    POSITIVE LOGITS
    æĭį
    0.11
    ä¸ĢæĿ¡
    0.10
    çĵ®
    0.09
    Unt
    0.08
    å¿ĥæĻº
    0.08
    åıijæĮ¥ä½ľçĶ¨
    0.08
    åĭĥ
    0.08
    åĵ©
    0.08
    ants
    0.08
    nings
    0.08
    Activations Density 0.077%

    No Known Activations