© 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. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 16311
    Prev
    Next
    INDEX
    Explanations

    frequency adverbs

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

    This neuron responds to first-person self-references, i.e. personal pronouns and phrases like “I,” “my,” “we,” or “our.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     now
    -1.24
    iſter
    -1.10
    ttps
    -1.04
     şimdi
    -1.04
     NOW
    -1.03
     Dänemark
    -1.03
     мл
    -1.02
     épices
    -1.00
     yet
    -0.98
     already
    -0.97
    POSITIVE LOGITS
     sometimes
    1.84
     always
    1.67
     всегда
    1.66
     my
    1.62
     often
    1.55
     иногда
    1.54
    sometimes
    1.54
    Sometimes
    1.46
    往往
    1.46
     usually
    1.45
    Activations Density 0.017%

    No Known Activations