© 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. 71525
    Prev
    Next
    INDEX
    Explanations

    sections and descriptions

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

    The neuron is specialized for detecting numeric tokens—years, section numbers, counts, and other numeric references in the text.

    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
     DIRECTORS
    -0.91
    gridColumn
    -0.91
     són
    -0.88
     영화
    -0.85
    imų
    -0.83
    tése
    -0.83
    ≿
    -0.83
    correre
    -0.79
    例えば
    -0.79
    cié
    -0.78
    POSITIVE LOGITS
     Overview
    1.11
     characteristics
    0.95
     overview
    0.95
    DESCRIPTION
    0.94
     how
    0.91
     Facts
    0.89
     principales
    0.88
     Description
    0.84
     How
    0.83
     Characteristics
    0.83
    Activations Density 0.320%

    No Known Activations