© 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. 22-GEMMASCOPE-RES-131K
    4. 95500
    Prev
    Next
    INDEX
    Explanations

    first

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

    The neuron fires strongly on content words—open-class terms like nouns, verbs, adjectives, and adverbs—while ignoring function words and punctuation.

    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
     when
    -2.52
     during
    -2.14
     after
    -2.03
     this
    -1.80
     while
    -1.53
     cuando
    -1.52
     now
    -1.52
     in
    -1.50
     since
    -1.50
     throughout
    -1.44
    POSITIVE LOGITS
     it
    1.56
    immediately
    1.38
     zuerst
    1.20
     hacerlo
    1.18
     flera
    1.18
     först
    1.17
     pierwszej
    1.16
     refroid
    1.09
     сначала
    1.09
     premières
    1.07
    Activations Density 0.222%

    No Known Activations