© 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. 84081
    Prev
    Next
    INDEX
    Explanations

    risk management, responsibilities, ALRI

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

    The neuron is detecting numeric expressions (digits, numerical tokens, hyphens/periods as part of numbers, and related numeric formatting).

    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
     Tampoco
    -1.26
     svært
    -1.25
     sensaciones
    -1.22
     работают
    -1.20
    🧆
    -1.17
     negociaciones
    -1.14
    zzling
    -1.13
     kovo
    -1.13
    摇了
    -1.11
    敕
    -1.11
    POSITIVE LOGITS
    <bos>
    1.10
     läs
    1.06
     bergen
    1.02
     Enkel
    1.02
     Heilige
    0.99
     algas
    0.99
    liferay
    0.96
     kronor
    0.96
    พาะ
    0.94
     Heilig
    0.94
    Activations Density 0.026%

    No Known Activations