© 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. 113033
    Prev
    Next
    INDEX
    Explanations

    federal reporter citations

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

    The neuron is picking up the “F” in Federal Reporter citations (e.g. the “F.” in “326 F.3d” or “320 F.2d”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    capital “F” used as an abbreviation or leading initial in legal citations and acronyms.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    the letter "F" when it appears in legal case citations (e.g., "F.2d", "F.3d") indicating Federal Reporter volumes.

    oai_token-act-pair · claude-4-5-sonnetTriggered 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
    曁
    -3.02
    鋮
    -2.78
    垍
    -2.70
    梟
    -2.64
    輓
    -2.59
     voor
    -2.56
    al
    -2.55
    踅
    -2.38
    ↵
    -2.36
     eléctrica
    -2.31
    POSITIVE LOGITS
    1
    3.42
    缡
    3.02
    2
    2.70
    źdz
    2.63
     шесть
    2.58
    莇
    2.55
    ANIEL
    2.48
    蓐
    2.44
    ’
    2.41
     пять
    2.39
    Activations Density 0.040%

    No Known Activations