© 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. 15802
    Prev
    Next
    INDEX
    Explanations

    symbols in code

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

    The neuron fires on the quoted punctuation sequences used in JSON (e.g. the ‘":"’ or ‘","’ separators between keys and values).

    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
    *
    -2.80
    "
    -2.61
    replyreport
    -2.58
     s
    -2.53
     –
    -2.50
    \
    -2.48
     ;
    -2.44
    </strong>
    -2.31
     extremadamente
    -2.30
     «
    -2.25
    POSITIVE LOGITS
     internacionais
    3.08
    鮡
    2.88
    ⌵
    2.80
    ꪹ
    2.78
    菪
    2.75
    zembro
    2.66
    cloths
    2.66
     producen
    2.59
     essenciais
    2.55
     値下げ
    2.50
    Activations Density 0.007%

    No Known Activations