© 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
    2. Qwen3-32B
    3. 32-RESID-BATCHTOPK-65K
    4. 5577
    Prev
    Next
    INDEX
    Explanations

    -eyed, weary, experience

    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
    ting
    -0.10
    ç±į
    -0.09
    æĻ°
    -0.09
    éªŀ
    -0.09
     orient
    -0.09
    çĴĺ
    -0.08
    -Clause
    -0.08
    èĬ¾
    -0.08
    æįħ
    -0.08
    -produ
    -0.08
    POSITIVE LOGITS
    ingly
    0.14
       
    0.11
    amp
    0.10
    iful
    0.10
    -eyed
    0.09
    ä¸Ģçľ¼
    0.09
    ãĥ£
    0.09
    hall
    0.09
    sburg
    0.09
     dÃłng
    0.09
    Activations Density 0.006%

    No Known Activations