© 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. 36665
    Prev
    Next
    INDEX
    Explanations

    launching or taking off

    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
    ãĥ¯ãĤ¤
    -0.09
    好äºĭ
    -0.09
    processors
    -0.08
    临港
    -0.08
    转移
    -0.08
    æĬĢ
    -0.08
    å·§åħĭåĬĽ
    -0.08
    rail
    -0.08
    ä½ŀ
    -0.08
    ç¡IJ
    -0.08
    POSITIVE LOGITS
    èµ·é£ŀ
    0.16
     launch
    0.13
     launches
    0.13
     Launch
    0.12
    launch
    0.12
    èħ¾é£ŀ
    0.12
    пÑĥÑģ
    0.12
     launching
    0.11
    Launch
    0.11
    Launching
    0.11
    Activations Density 0.090%

    No Known Activations