© 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. 27337
    Prev
    Next
    INDEX
    Explanations

    forms of 'touch'

    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
    åľºä¸Ĭ
    -0.09
    ilers
    -0.09
    ä¸Ģéĥ¨
    -0.08
    gere
    -0.08
    æĸ¯å¡Ķ
    -0.08
    ä¸Ńå°ıåŀĭ
    -0.08
    å¼ĢæĶ¾å¼ı
    -0.08
    丰满
    -0.08
    .Diff
    -0.08
     pur
    -0.08
    POSITIVE LOGITS
    touches
    0.14
    -touch
    0.14
     touches
    0.13
     touch
    0.13
    (touch
    0.13
     touched
    0.12
    .touch
    0.12
    触æij¸
    0.12
    touch
    0.12
     TOUCH
    0.11
    Activations Density 0.017%

    No Known Activations