© 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. 34-GEMMASCOPE-RES-131K
    7. 90463
    Prev
    Next
    INDEX
    Explanations

    claiming wins or credit

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

    This neuron activates for markers of reported speech or quotations—i.e. words and punctuation used to attribute statements (like “claimed,” “told,” question‐marks ending quotes, etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     alright
    -0.92
    Alright
    -0.92
    According
    -0.85
     vitale
    -0.82
     esportes
    -0.77
     intellectuelle
    -0.76
    确保
    -0.75
     blessed
    -0.74
    gato
    -0.74
     Alright
    -0.72
    POSITIVE LOGITS
     Us
    0.87
     cof
    0.84
    Vri
    0.82
    Rarity
    0.81
     Latvia
    0.81
    iable
    0.80
    trs
    0.79
    casecmp
    0.79
    cokinetic
    0.79
     diversa
    0.78
    Activations Density 0.003%

    No Known Activations