© 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. 7165
    Prev
    Next
    INDEX
    Explanations

    website navigation and content terms

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

    The neuron fires on words in user‐interface or help text—i.e. the instructional/action language telling you what to do (enable, play, download, post, reserve, report, etc.) and the objects of those actions.

    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
     to
    -1.64
     by
    -1.02
    Here
    -1.01
     particolarmente
    -0.83
    chures
    -0.83
     haram
    -0.79
     desvan
    -0.79
     ederim
    -0.79
    ทำให้
    -0.78
     вред
    -0.78
    POSITIVE LOGITS
     ולה
    1.20
    ्ल
    0.96
    fat
    0.94
    UpperCase
    0.92
    meyi
    0.89
     kiệm
    0.87
    fan
    0.86
    dar
    0.85
     ול
    0.85
     offseason
    0.84
    Activations Density 0.138%

    No Known Activations