© 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. 22-GEMMASCOPE-RES-131K
    7. 98157
    Prev
    Next
    INDEX
    Explanations

    list item separator

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

    The neuron responds to tokens that are composed of—or contain—uppercase letter sequences (i.e. acronyms, all-caps words, or uppercase identifiers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    ᕮ
    -1.09
     equipes
    -1.05
    
    -1.04
     Selección
    -1.02
     trám
    -1.01
     maniere
    -1.01
     miſ
    -1.01
    ౧
    -1.00
    whose
    -0.99
    decided
    -0.99
    POSITIVE LOGITS
     highly
    1.28
     very
    1.07
     slight
    1.05
     those
    1.05
     similar
    1.02
     mostly
    1.02
     if
    1.00
     will
    0.99
     while
    0.99
     greatly
    0.99
    Activations Density 0.132%

    No Known Activations