© 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. 10-GEMMASCOPE-RES-131K
    7. 69380
    Prev
    Next
    INDEX
    Explanations

    in due course, in honesty, in likelihood, at last

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

    The neuron activates on the key nouns in idiomatic adverbial or discourse‐marker phrases (e.g. “in all honesty,” “in due course,” “at long last,” “in all likelihood,” etc.).

    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
     leiding
    -2.30
    "$_
    -2.06
    老爷子
    -2.00
    驍
    -1.80
     experimentado
    -1.80
    飚
    -1.78
    🧧
    -1.77
    書いた
    -1.74
     zilver
    -1.73
     groter
    -1.73
    POSITIVE LOGITS
    一个
    2.63
    6
    2.36
    k
    2.23
    b
    2.22
    1
    2.20
    要
    2.19
    ll
    2.09
    4
    2.02
    v
    2.00
    \
    1.98
    Activations Density 0.012%

    No Known Activations