© 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. 13874
    Prev
    Next
    INDEX
    Explanations

    function calls with parentheses

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

    The neuron flags content‐bearing, uncommon tokens (e.g. identifiers, rare words, or literal strings) rather than common stopwords or 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
     mädchen
    -1.17
    dacia
    -1.09
     BISHOP
    -1.07
    Martín
    -1.07
    resumen
    -1.07
     рассказала
    -1.07
     jahr
    -1.06
    ulsión
    -1.05
    的关键
    -1.05
    式的
    -1.05
    POSITIVE LOGITS
     sowie
    1.34
    but
    1.13
     “...
    1.09
     النبات
    1.05
    ™,
    1.04
    ®,
    1.04
    ("
    1.03
    𝗦
    1.02
    嚓
    1.02
     (“
    1.01
    Activations Density 0.056%

    No Known Activations