© 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. 12786
    Prev
    Next
    INDEX
    Explanations

    action verbs and specific language

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

    The neuron fires on words ending in “-ing” that denote actions or processes (gerund/present-participle forms), especially in technical or methodological contexts.

    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
    z
    -2.58
    e
    -2.31
    y
    -2.27
    j
    -2.19
    p
    -2.16
    s
    -2.14
    q
    -2.08
     его
    -2.05
    t
    -2.00
    “
    -1.95
    POSITIVE LOGITS
     the
    2.72
     a
    2.38
     parachoque
    2.16
    盦
    2.05
     esqueleto
    2.03
    欵
    1.98
    へ行
    1.97
    おわりに
    1.97
     espet
    1.96
    酲
    1.96
    Activations Density 0.235%

    No Known Activations