© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 34531
    Prev
    Next
    INDEX
    Explanations

    morning greetings and names

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

    This neuron detects message-level salutations and header metadata (e.g. “Hi,”, “To:”, timestamps) that appear at the starts of emails or chat messages.

    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
    mbangkan
    -1.16
    Training
    -0.92
    женных
    -0.84
    Journey
    -0.82
    ография
    -0.81
     conoz
    -0.81
    respondent
    -0.81
     писать
    -0.80
     bambou
    -0.80
    Tutorial
    -0.80
    POSITIVE LOGITS
     you
    0.83
    HeaderComponent
    0.80
    FACEBOOK
    0.79
    を行います
    0.79
    也得
    0.78
     пусть
    0.78
    のだろう
    0.76
     electr
    0.75
     nutr
    0.73
    belli
    0.73
    Activations Density 0.034%

    No Known Activations