© 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. 10-GEMMASCOPE-RES-131K
    4. 11543
    Prev
    Next
    INDEX
    Explanations

    upcoming specific events and dates

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

    The neuron strongly activates on numerical or date/time information (e.g. tokens like “for” before numbers, standalone numbers, dates, and time indicators).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     instead
    -1.27
     then
    -1.25
     Instead
    -1.20
     June
    -1.16
     biasanya
    -1.14
     have
    -1.10
    крыть
    -1.09
    八十
    -1.05
    主意
    -1.05
     plutôt
    -1.05
    POSITIVE LOGITS
    内核
    1.10
    鋦
    1.07
     conseille
    1.06
     environs
    1.05
     voul
    1.04
     weet
    1.04
     informasjon
    1.04
     ijs
    1.04
     cerveja
    1.03
     BURN
    1.03
    Activations Density 0.191%

    No Known Activations