© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 13269
    Prev
    Next
    INDEX
    Explanations

    numbers and code snippets

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

    The neuron fires on tokens involved in exponentiation and other arithmetic‐operator notation (e.g. “**”, “*”, numbers used in power expressions).

    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
     plán
    -1.10
     céramique
    -1.03
     lèvres
    -1.02
     ),
    -1.02
     fanns
    -1.00
     góry
    -0.97
     Aussehen
    -0.94
     estante
    -0.93
     Parsing
    -0.93
     Emits
    -0.92
    POSITIVE LOGITS
     –
    0.97
    Ữ
    0.96
     乐
    0.96
     传统
    0.96
     stund
    0.96
     压
    0.87
     domestiques
    0.85
     WaitForSeconds
    0.84
    ledad
    0.84
    ERS
    0.83
    Activations Density 0.340%

    No Known Activations