© 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. 11613
    Prev
    Next
    INDEX
    Explanations

    EFI, Edo, LPG, anticoag

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

    acronyms and technical terminology, particularly "EFI", "EDO", "ORM", and "LPG" in technical documents.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron activates on multi‐letter uppercase abbreviations (technical acronyms) such as EFI, UEFI, LPG, ORM, 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
    大腿
    -2.69
    tafogo
    -2.45
     gustaMe
    -2.42
    媵
    -2.19
     dieß
    -2.16
     the
    -2.09
     dames
    -2.09
    訁
    -2.08
    ܃
    -2.05
     が
    -2.03
    POSITIVE LOGITS
    </strong>
    2.88
     karena
    2.70
    ↵↵
    2.63
    Y
    2.45
            
    2.41
    S
    2.36
    <em>
    2.34
     heighten
    2.31
    E
    2.31
     在
    2.23
    Activations Density 0.009%

    No Known Activations