© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 45740
    Prev
    Next
    INDEX
    Explanations

    Haq, Hak, khaki

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron activates on any token containing the letter sequence “hak” (e.g. in names like Haq or Hakim or words like khaki).

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     nová
    -2.36
    廔
    -2.30
    線が
    -2.30
    我们在
    -2.23
    籟
    -2.17
     欣赏
    -2.17
     articulate
    -2.17
    鸶
    -2.16
     Then
    -2.11
    大量的
    -2.11
    POSITIVE LOGITS
    4
    2.81
    0
    2.59
    and
    2.11
    5
    2.09
    In
    2.09
    毟
    2.08
    明日は
    2.06
    On
    2.02
    6
    2.02
    for
    1.97
    Activations Density 0.007%

    No Known Activations