© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 128220
    Prev
    Next
    INDEX
    Explanations

    FG prefix or color attribute

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

    The neuron activates on uppercase “FG” sequences in technical acronyms (e.g. FGF, FGFR, FG-PET), i.e. biomedical gene/protein or imaging abbreviations starting with “FG.”

    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
    ).
    -3.19
    ↵
    -3.17
    e
    -3.08
    ทั้ง
    -2.67
    </b>
    -2.63
    杙
    -2.53
     stuffy
    -2.44
    a
    -2.31
    すっかり
    -2.17
     –
    -2.11
    POSITIVE LOGITS
    釭
    2.69
     selben
    2.64
    msford
    2.56
     enfati
    2.45
    ?"
    2.45
    >
    2.44
     terceiros
    2.38
    彇
    2.36
    婬
    2.36
     vorigen
    2.34
    Activations Density 0.005%

    No Known Activations