© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 130902
    Prev
    Next
    INDEX
    Explanations

    message IDs and references

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

    This neuron detects occurrences of the term “message” (including its abbreviation “msg”), especially in contexts of message fields or identifiers.

    oai_token-act-pair · o4-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
    },
    -1.31
     xt
    -1.20
     خلال
    -1.20
     diseñados
    -1.16
     for
    -1.15
    ↵↵
    -1.15
     Bà
    -1.14
     seepage
    -1.12
     &
    -1.07
    isan
    -1.07
    POSITIVE LOGITS
     your
    1.53
     about
    1.45
     isolada
    1.23
     饰
    1.22
     recomendada
    1.20
     estrés
    1.18
     them
    1.16
     any
    1.16
    صاص
    1.15
     recent
    1.12
    Activations Density 0.052%

    No Known Activations