© 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. 48007
    Prev
    Next
    INDEX
    Explanations

    recognizing abbreviations and their meanings

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

    The neuron activates on explicit mentions of “abbreviation” (and its variants like “abbreviations”), especially in headings or lists defining shortened forms.

    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
    )
    -2.89
     hinder
    -2.41
     sagte
    -2.39
    蓠
    -2.34
     podem
    -2.31
    lysses
    -2.27
    mberto
    -2.23
    榖
    -2.23
    ſu
    -2.16
    zaSy
    -2.16
    POSITIVE LOGITS
    酲
    3.41
    ↵
    2.92
    ～。
    2.83
    7
    2.64
    ～！
    2.61
    1
    2.59
    ⽌
    2.48
    猡
    2.44
    疍
    2.42
    淠
    2.33
    Activations Density 0.009%

    No Known Activations