© 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. 105271
    Prev
    Next
    INDEX
    Explanations

    superlatives and their follow-ups

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

    The neuron spots superlative or high-degree adjectives (e.g. “most,” “greatest,” “best,” “ultimate,” “recent,” “useful”).

    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
     you
    -2.59
     of
    -2.13
     there
    -1.89
     they
    -1.76
    必要的
    -1.49
    crema
    -1.44
     jeder
    -1.42
    寒冷
    -1.42
     бъ
    -1.41
    ࿊
    -1.41
    POSITIVE LOGITS
     own
    2.08
     and
    1.65
    躋
    1.49
    ",
    1.47
    大约
    1.45
     hoeft
    1.44
     skvě
    1.40
    </i>
    1.40
    3
    1.40
    大約
    1.38
    Activations Density 0.044%

    No Known Activations