© 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. 72782
    Prev
    Next
    INDEX
    Explanations

    really + adjective/adverb

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

    The neuron fires on emphatic or intensifying words (e.g. “really,” “this”) that signal strong emphasis or assertion.

    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
     tujuh
    -1.82
    име
    -1.80
     of
    -1.80
    ейчас
    -1.80
    
    -1.77
    汌
    -1.75
    卝
    -1.74
    珦
    -1.73
    戋
    -1.70
    紘
    -1.70
    POSITIVE LOGITS
    1
    2.09
    ↵
    1.84
    T
    1.64
    十几
    1.59
     vitality
    1.53
     Including
    1.50
    これで
    1.50
     Appears
    1.45
    <i>
    1.38
    M
    1.38
    Activations Density 0.002%

    No Known Activations