© 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. 57629
    Prev
    Next
    INDEX
    Explanations

    words indicating quantity

    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
    èĢĹ
    -0.10
    ä¸Ģå¥Ĺ
    -0.10
    éģ£
    -0.10
    ä¼ĹæīĢ
    -0.09
    _secs
    -0.09
    ORA
    -0.09
    äºĮåŃĹ
    -0.08
    endas
    -0.08
    ãģĦãģĭ
    -0.08
    çĭ¸
    -0.08
    POSITIVE LOGITS
     them
    0.09
    æ³±
    0.09
    odd
    0.08
    ãģ¿ãĤĵãģª
    0.08
    å®ĺ
    0.08
    åıĤä¸İèĢħ
    0.08
    浩
    0.08
    è¯ĦåĪ¤
    0.08
    è¿ĻäºĽäºº
    0.08
    纷纷
    0.08
    Activations Density 0.124%

    No Known Activations